Journal of Marine Research, 41, 215-237,1983
Light-shade adaptation and vertical mixing
of marine phytoplankton: A comparative field study
by Paul G. Falkowski1
ABSTRACT
The hypothesis is examined that the recent light history of phytoplankton contains information about vertical mixing processes in the euphotic zone. Chlorophyll/P7oo ratios are used to
estimate the degree of light or shade adaptation in natural phytoplankton communities. Along
with information about the time- and light-dependent
rates of change of cWorophyll/P7oo
ratios, a model is presented to estimate how recently populations at the surface were at the 1 %
light depth and vice versa. The model is based on first-order kinetics and employs a temperature correction. The model is used to estimate vertical displacement rates (i.e., piston velocities)
on Georges Bank, in the New York Bight, and off the coast of Hawaii. The results suggest that
vertical displacement rates vary by about two orders of magnitude (from ca 3.8 x 10-0 em/sec
to 1.1 X 10-1 em/sec). These values are in general agreement with theoretical calculations
based on physical parameters.
1. IutrodUctiOD
In phytoplankton dynamics the importance of vertical mixing processes for the
availability of nutrients, light, and the distribution of biomass is qualitatively known
(e.g., Sverdrup, 1953; Riley, 1965). Because of the difficulty in directly measuring
vertical velocities, however, little progress has been made in quantitatively relating
vertical mixing to phytoplankton production and distribution. Vertical eddy diffusivities can be estimated from vertical nutrient gradients and phytoplankton nutrient
demands (Eppley et aI., 1979; King and Devol, 1979); however, this approach cannot be used unless a nutricline is a prominent characteristic of the system. In this
study, the hypothesis is presented that the recent light history of phytoplankton may
provide information about vertical mixing processes.
. Let us consider the light regimes experienced by marine phytoplankton in natural
waters. First-order variations in the light regimes are influenced by the path of the
sun (i.e., the diurnal photoperiod). Diurnal variations in light can be temporally
manifested in such properties as the diel periodicity of carboni chlorophyll ratios
1. Oceanographic Sciences Division, Department of Energy and Environment, Brookhaven National
Laboratory, Upton, New York, 11973, U.S.A.
215
216
Journal of Marine Research
[41, 2
(Owens et aI., 1980), or photosynthetic assimilation numbers (Malone, 1971). Second-order variations in light regimes are primarily influenced by the vertical displacement of cells in the water column (Falkowski and Wirick, 1981). Second-order
variations in light are spatially manifested in vertical variations in biochemical or
physiological components of the cells. From these premises it can be inferred that
vertical variations in biochemical or physiological components can be related to
vertical displacement rates in situ.
It is recognized that the light history of a phytoplankton community influences the empirical relationship between photosynthesis and
irradiance (Steemann Nielsen and Hansen, 1959; Ryther and Menzel, 1959; Steemann Nielsen and Park, 1964; Beardall and Morris, 1976; Marra, 1978; Falkowski,
1980). The empirical relationship between photosynthesis and irradiance is mediated
by physiological and biochemical adaptations, generally referred to as light-shade
adaptation (Prezelin and Sweeney, 1979; Falkowski, 1980; Falkowski and Owens,
1980; Perry et al., 1981). Light-shade adaptation is characterized by changes in
photosynthetic pigment content and photosynthetic response, and is often accompaIlied by changes in chemical composition, fluorescence characteristics, and cell
volume (Myers and Graham, 1971; Falkowski and Owens, 1980).
Consider a vertical mixing process that displaces a cell on some time scale longer
than the time it takes for the cell to become adapted to the difference in light regimes
between depths Zo and Zl. If the cells maIlifest some characteristic of adaptation to
light, vertical variations in this characteristic would reflect the light history of the
cell at depth Zo, and would be nonuniformly distributed vertically because of a new
characteristic associated with the light regime at depth Zl' If the time scale of change
in the adaptive variable is known, it is possible to estimate the maximum rate of
vertical displacement of the cell in the water column. If mixing processes occur on
a time scale shorter than the time it takes for the cells to adapt to the variations in
the light regime, the vertical distribution of the light-dependent physiological characteristic would be expected to be more uniformly distributed. From knowledge of
the time scale of change of the adaptive variable, a minimum vertical mixing rate
can be estimated. Implicit in this hypothesis is that second-order variations in lightdependent physiological attributes are not masked by first-order variations.
One method that may be used to estimate vertical displacement rates is toestimate the recent light history of phytoplankton by generating P vs I curves from
samples taken at known light depths (Steemann Nielsen and Hansen, 1961; Eppley
and Brooks, unpub!. ms.). The assumption may be made that the differences in such
photosynthetic parameters as the initial slopes of P vs I curves (i.e., the light utilization efficiency), It, values, or Pmax values reflect the light history of the sample. For
example, differences in Pmax or It, values between samples collected at depths z and
z ± Az might be used to estimate the degree to which two vertically separated phytaTheoretical
considerations.
Falkowski: Adaptation and mixing
1983J
217
plankton populations are adapted to high or low light intensities. H kinetic coefficients of light-shade adaptation (vis-a-vis Pmax or
values) are known, it is possible
to estimate how far apart in time spatially separated samples are and thereby estimate an upper or lower bound for a vertical displacement rate.
'This type of approach was taken by Steemann Nielsen and Hansen (1959) and
Ryther and Menzel (1959) to qualitatively estimate the degree of vertical mixing.
Eppley and Brooks (unpubl. ms.) attempted to semiquantitatively estimate mixing,
using laboratory-derived coefficients for the time- and light-dependent rates of
change in Ik values. The validity of such estimations of vertical mixing rates, using
P vs-I parameters, depends upon (among other things) the length of the incubation
period with radiocarbon. In some species (e.g., Skeletonema costatum), light-shade
adaptation may occur within a few hours (JllSrgensen,1969; Riper et aI., 1979), so
that the P vs I parameters measured may not reflect the physiological state of the
cells at the time of collection.
Changes in such photosynthetic parameters as light utilization efficiency, P max, or
values are thought to be brought about by changes in the size aJld/ or numbers of
photosynthetic units (ct. Herron and Mauzerall, 1971; Prezelin and Sweeney, 1979;
Falkowski, 1980; Falkowski et aI., 1981). This premise frames another approach
that may be used to estimate the light history of phytoplankton. It is hypothesized
that as cells become more shade adapted, the size of photosynthetic units increases.
This hypothesis is supported by numerous laboratory studies on the characteristics
of light- and shade-adapted unicellular algae (Myers and Graham, 1971; Falkowski
and Owens, 1980; Falkowski et aI., 1981; Perry et aI., 1981).
A number of methods have been described for measuring the size of photosynthetic units (pSU) (Falkowski et al., 1980). In this study, photosynthetic unit (PSU)
sizes were estimated by measuring the ratio of chlorophyll/P 100 using a spectrophotometric technique (e.g., Shiozawa et al., 1974; Falkowski and Owens, 1980).
A diagram outlining the principle of this approach is given in Figure 1. As in the
use of
values to estimate light history, the ratio of photosynthetic unit sizes between two depths can be used as a relative index of the recent light history of a
phytoplankton community within a given water column. This technique requires no
incubation period, thereby preventing cells from adapting to a new light regime.
Using both chl/P100 ratios and P vs I curves to estimate the light history of phytoplankton, a model is presented for the estimation of upper and lower bounds of
vertical displacement rates in three contrasting regions: the southern portion of
Georges Bank (GB), the New York Bight (NYB), and the subtropical Pacific (SP)
at the site of a proposed OTEC plant off the coast of Hawaii. As a result of tidal
mixing, GB is never stratified (Hopkins and Garfield, 1981); the NYB is vertically
mixed in the winter but becomes intensely stratified in the summer (Bumpus, 1973;
Beardsley et al., 1976); in the SP a permanent deep thermocline is present below
the euphotic zone, and a shallow one develops seasonally.
I,.
I,.
I,.
[41, 2
lournal of Marine Research
218
SCHEMATIC MODEL OF PSI
LIGHT HARVESTING TRAP
LIGHT
IN LIGHT ADAPTED CELLS
Chi Proo
Chi a I Proo- LOW
IN SHADE ADAPTED CELLS'
Chi Proo
ChIlli
Proo-H1GH
Figure 1. Schematic representation of the changes that occur in a photosynthetic unit as a
typical diatom undergoes light-shade adaptation. The photosynthetic unit may be cOnceived
of as a dish antenna, the collecting area containing chlorophylls and accessory pigments. At
the focus of the antenna is a reaction center-here
represented by .a P••• molecule. As the
cell becomes shade adapted the antenna size increases, leading to a proportional increase hi
chlorophyll/P,"" ratios. When shade-adapted cells are exposed to high light, the reverse
phenomenon occurs (cf. Falkowski and Owens, 1980).
2. Materials and methods
Data used in this study were collected from five cruises between 1977 and 1979
(Table 1). Each day a productivity station was occupied at ca 0800 h local time. At
these stations, light extinction coefficients were determined with a Lambda LI 1925
quantum sensor interfaced with a ratio amplifier. A second sensor, placed on the
deck, served as a reference. The ratios of surface light (1 to light at known depths
0)
Table 1. List of cruises, regions, and times when chl/P,"" ratios were measured.
Vessel
Region
Month
R/V
R/V
R/V
R/V
R/V
NYB
NYB
NYB,GB
NYB, GB
SP
March 1977
August 1977
April 1978
October 1978
October 1979
Kelez
Knorr
Atlantis 11
Atlantis 11
Kona Keoki
1983]
Falkowski: Adaptation and mixing
219
(I,,) were output directly onto a recorder. Log 1,,/10 values were plotted as a function
of depth and extrapolated through the origin. Sampling depths were determined
from the light extinction plots and corresponded tol00, 60, 30, 15,5, 1, and 0.1 %
of I".
Samples were obtained with Niskin bottles. For P vs I curves, samples were drawn
from selected light depths (usually 100 and 1%) and incubated with 5 f'Ci of
NaH14C03 in screened bottles at 100, 60, 30, 15, 5, and 1% light. A dark bottle
was run in parallel for each depth. These samples were incubated for 2 h in fluorescent light incubators at in situ temperatures. Temperatures were maintained with
Neslab CFf 75 recirculating water baths. The sample bottles were mechanically
rotated in the incubators at 6 rpm, and light was provided from two sides by VHO
fluorescent tubes supplying 675 p,E m-2 sec-1 (PAR).
Upon completion of the incubations, the samples were immediately filtered onto
Millipore HA filters at 10 mm Hg vacuum pressure, fumed for 60 sec over concentrated HCl, and placed in scintillation vials to which 10 ml of Filter Solv (Beckman)
was added. The samples were counted in a Beckman LS 3150T scintillation counter
with an external standard. Dark bottle counts were subtracted from the corresponding light bottle counts in the productivity calculations (Strickland and Parson, 1972).
Chlorophyll a was determined fluorometrically from 90% acetone extracts.
Freshly drawn samples were filtered on Reeve Angel 984H glass fiber filters and
ground in a tissue grinder with 90% acetone. The acetone extracts were filtered into
clean graduated centrifuge tubes to remove the glass fibers, brought to a constant
volume, and measured before and after acidification in a Turner Designs Model 10
fluorometer (Yentsch and Menzel, 1963) calibrated with pure chlorophyll a.
In situ measurements of in vivo chlorophyll fluorescence were made with a Turner
Designs Modell 0 fluorometer housed in a watertight pressure case. This instrument
was also equipped with a thermistor and depth transducer, so that fluorescence,
temperature, and depth (FTD) could be simultaneously measured.
To measure PSU unit sizes, 4- to 60-1 samples were taken from selected depths
with Niskin bottles (Fig. 2). The amount of water processed depended on the chlorophyll a concentrations at the desired depths. Samples were either filtered directly
on 4.7-cm Gelman AE glass filters (10 1 or less) or continuously centrifuged to a
small volume (ca 100 ml) and then filtered. Filtration or centrifugation was completed within 1 h of collection. The filters were homogenized under dim light in 2 ml
of 50 mM ice-cold Tris-HCl buffer, pH 7.8, containing 0.01 % Triton X-I00. The
suspensions, containing crude chloroplast membranes, were centrifuged at 2000
x g for 5 min to remove glass fibers and large cell debris. The supernatants were
scanned with an opal glass filter (Shibata, 1958) and beam scrambler from 350 to
750 nm in the split beam mode of Aminco DW-2a spectrophotometer against distilled water. Chlorophyll a was determined in the aqueous suspensions by subtracting the absorbance at 750 nm from that at 678 nm and using an absorptivity coeffi-
220
[41, 2
lournal of Marine Reuarch
4 - 60 I SAMPLE
COLLECTED 8Y NISKIN
BOTTLE FROM 01 SCRErE
DEPTHS
SAMPLES FILTERED DIRECTLY
ON 47mm GLASS FIBER FILTERS
OR
CENTRIFUGED CONTINUOUSLY
AND THEN· THE PELLETS
FILTERED
HOMOGENIZE FILTER
WITH 2ml OF·50mM
TRIS-HC I (pH7.8)
CONTAINING 0.01-'_
TRITON X-loo
(0-4-C)
SCAN 350 TO 700nm IN
SPLIT BEAM MODE OF AMI NCO
DW-20
SPECTROPHOTOMETER
MEASURE OXIDIZED !!!!tl!1
REDUCED P700 IN DUAL
8EAM MODE OF AMINCO OW-20
SPECTROPHOTOMETER
TIME
Figure 2. Flow diagram for the processing and measurement of chlorophyll/P"", ratios in the
field.
dent of 60 mM-l cm-l (Thornber et aI., 1977). Solid methyl viologen and sodium
ascorbate were added to the sample to final concentrations of ca 100 pM and
10 mM, respectively, and P700 was measured in a 10 X 4-mm cuvette in the dual
wavelength mode of the spectrophotometer by following the light-induced absorption changes at 697 nm relative to the isosbestic wavelength of 720 nm (Falkowski
and Owens, 1980; Hiyama and Ke, 1972). Actinic illumination of S sec duration was
provided by a focused ISO-watt tungsten source filtered through two Corning 5543
filters (A.max = 420 nm). The photomultiplier was protected by a single Corning
2030 blocking filter. Samples were held in the secondary sample position to eliminate
chlorophyll fluorescence. Care was taken to ensure that the actinic source was suf·
1983]
Falkowski: Adaptation and mixing
221
ficiently bright to achieve light saturation of P100 photochemical activity. P100 signals
were averaged over 10 flashes for each sample. P100 was calculated from the lightinduced absorbance change using a difference absorptivity coefficient of 64 mM-l
cm-l (Shiozawa et al., 1974). Typical absorption differences arising from the P10D
signal are small (ca 5 x 10-· o.d. units), requiring a high-sensitivity spectrophotometer; the coefficient of variation of the technique is ± 14%. The molar ratios of
chlorophyll a and P100 were used to estimate the average size of PSUs from each of
the sample depths.
Temperature was measured with reversing thermometers, salinity with a Guildline salinometer, and nutrients (NOa-, N02-, NH.+, PO.2-, Si(OH).) with an
AutoAnalyser (Whitledge et aI., 1981). Phytoplankton samples were taken from
desired depths, preserved with Lugol's iodine, and counted in Ottermohl chambers
with an inverted microscope (Falkowski and von Bock, 1979).
3. Results and discussion
In 1960, Ryther and Hulburt suggested that a mixed layer as inferred from physical or chemical criteria is not necessarily mixed with respect to phytoplankton. This
apparent anomaly is primarily related to two facts: (a) the density field does not
provide information about the energy available for mixing at the time of observation,
and (b) the distribution of phytoplankton is nonconservative. This principle is illustrated with a few specific examples.
Example 1. Consider two vertical profiles taken in the spring: one in April from
the southern region of GB (Fig. 3a), the other in March from a similar isobath in
the NYB (Fig. 4b). On GB both the density and chlorophyll fields are uniformly
distributed with depth, suggesting that vertical mixing occurs on a relatively rapid
time scale. Tidal stirring continuously provides enough energy and vertically distributes heat (Hopkins and Garfield, 1981) and chlorophyll such that vertical
gradients in density or phytoplankton biomass do not form. At a similar isobath in
the NYB, winter mixing processes are primarily driven by convective overturn and
irregular wind events. In the absence of recent wind events, the density profile in
this region is nonuniform, increasing by one (Tt unit over the 3()...mwater column.
It could be inferred from the vertical chlorophyll profile that little energy had been
recently available for mixing; i.e., the phytoplankton appear to sink out of the water
column faster than mixing processes redistribute them (Malone et aI., 1983).
Example 2. The onshore-offshore gradient in mixing energy in the NYB in the
spring is complicated by estuarine influences near the apex of the NYB and frontal
systems near the shelf break. In the apex of the NYB, nutrient-rich water from the
Hudson and adjoining estuaries is vertically separated from continental shelf water
by a halocline. The plume of this estuarine system is characterized by high standing
stocks of phytoplankton and high primary productivity (Malone, 1977). Vertical
222
[41, 2
lournal of Marine Research
fL9Chlo/1
,
8
12
I
fL9 chlo/I
16
0
I
I
2
I
CTt
CTt
0
24
-.5 2010
0
24
26
10
20
N
30
30
40
40
(a )
l b)
Figure 3. Vertical profiles of density (e) and chlorophyll (.~) OD southern Georges Bank in
(a) April 1978 and (b) October 1978. These profiles are taken at approximately the same
station location (-40°60' N, 68°04' W).
profiles of the density fields in this region (Fig. 4a) suggest that a weak mixed layer
extends to ca 8 to 10 m in March, about the same order of depth as the euphotic
zone. The chlorophyll profile suggests that enough mixing energy is available in the
upper mixed layer to evenly distribute phytoplankton in the euphotic zone, but not
throughout the entire water column. At the shelf~slope front, however, the water
column appears vertically stable (Fig. 4c) as a result of the interleaving of warm
slope water with cold shelf water. At this frontal region the vertical chlorophyll pr~
files often reveal higher phytoplankton biomass near or at the surface, implying that
growth rates exceed vertical mixing rates, at least within the upper 20 m (i.e., the
euphotic zone).
Example 3. The southern portion of GB is vertically uniform at anyone time and
vertical profiles of O't do not qualitatively change on seasonal time scales (Fig. 3a
and b). Tidal stirring provides sufficient energy to prevent formation of a pycnocline
in the warmer months, and the vertical chlorophyll distributions are qualitatively
similar throughout the year. Phytoplankton species composition throughout the water
column at any given time shows the same rank order abundance of floristic components (Falkowski and von Bock, 1979), which suggests that the vertical turnover
of the water column is as fast as or faster than the mean growth rate of the phytoplankton.
In contrast, in the summer in the midshelf of the NYB warm shelf water overlays
cold shelf water (the so-called "cold pool"; Bigelow, 1933; Ketchum and Corwin,
1964; Hopkins and Garfield, 1981), resulting in intense stratification (compare Fig.
4b and d). The turbulence in the upper mixed layer (between 5 and 10m thick) is
1983]
Falkowski: Adaptation and mixing
10
,
20
0
"'ge~I"/1
.
30
21
22
20
.0
,
I
223
"'ge~I"/1
50
I
crt
23
24
60
I
70
I
. . . . . . . .
80
2
4
6
8
to
12
14
cr ,
24
27
0
10
20
1,1
30
.
0
!
2
I
•
1"-ge~I,,/1
I
6
8
I
I
a;
N
24
10
12
I
I
27
1"-ge~1,,/1
2
0
I
I
022
I
vi 24
•
I
n°
10
20
30
40
50
60
Id I
70
80
90
100
lel
Figure 4. Vertical profiles of density (.) and chlorophyll (~) in the New York Bight. (a)
Near the mouth of the Hudson estuary in March 1977 (40°24' N, 73°53' W); (b) in the midshelf in March 1977 (30°39' N, 73°50' W); (c) at the shelf-break front in March 1977 (39°35'
N, 72°25' W); and (d) in the midshelf in August 1977 (39"46' N, 73°46' W).
usually sufficient to maintain a uniform distribution of chlorophyll above the pycnocline. This mixed layer is often shallower than the euphotic zone (Falkowski et al.,
1983). The pycnocline, a region of high vertical stability coupled with a high nutrient
gradient, provides a region conducive to the growth and accumulation of phytoplankton (d. Cullen and Eppley, 1981). A chlorophyll maximum, often coincident
with the depth of the pycnocline, is normally found in this region. As in the Southern
California Bight, the rank order abundance of phytoplankton species in the upper
mixed layer of the NYB is, more often than not, completely different from the
224
[41; 2
Journal of Marine Research
Table 2. Estimations of vertical displacement rates and turnover
Georges Bank, the New York Bight,. and the subtropical Pacific.
T'
Region
GB
(Southern)
GB
(Southern)
NYB
(Apex)
NYB
(Apex)
NYB
(Midshelf)
NYB
(Midshelf)
NYB
(Midshelf)
NYB
(Shelf break)
NYB
(Shelf break)
SP
(OTEC site)
(0C)
z••• z.··
der,···
dz (m-
of the mixed layer from
Chill min···
ChI/1 max
(Chl/P.oo)z. t
(m)
(m)
4.5
45
15
2.2 X 10-
0.93
725
13.0
45
22
1.0 X 10-
0.8.5
700
March 1977
4.1
7
7
2.5 X 10-1
0.84
400
August 1977
21.0
7
12
1.4 X 10-1
0.08
190
March 1977
3.0
35
21
4.3 X 10"'"
0.39
220
April 1978
3.7
50
30
4.1 X 10-
0 ..56
520
August 1977
24;2
12
16
2.1 X 10-'
0.07
510
March 1977
6.2
50
30
1.0 X 10-
0.84
360
August 1977
25.0
30
42
8.3 X 1~
0.25
310
October
27.0 -55
84
1.8 X 10-
0.47
110
Month
April 1978
October
1978
1979
I)
,
- average temperature in upper mixed layer.
• - depth of mixed layer .
•• - depth of euphotic zone.
••• - calculated over Z •.
phytoplankton assemblage found in the chlorophyll maximum (cf. Reid et al., 1978).
These results suggest that the vertical exchange between the upper mixed layer and
the pycnocline is slower than phytoplankton growth rates.
These and other examples of the lack of correlation between water column stability and mixing can be demonstrated by inspection of the data in Table 2. The
degree of stability is proportional to dUtldz while the degree of recent mixing can
be represented by an "evenness" coefficient of chlorophyll distribution; i.e., the ratio
of the maximum chlorophyll value to the minimum chlorophyll value in the euphotic
zone (cf. Cullen and Eppley, 1981). In four cases the chlorophyll evenness index
exceeds 0.8 (i.e., the water column is well mixed with respect to chlorophyll), however, these ratios span over four orders of magnitude of vertical stability.
Effects of vertical mixing on phytoplankton light history. The qualitative comparison of the vertical density and chlorophyll fields can serve as the basis of a
hypothesis: a comparison between the density field and some property of the chIaro-
1983]:
Falkowski: Adaptation and mixing
,.
(P", •• Jlk. ttt
(P •••u:JI)z:.
Estimated vertical
displacement rate
(cm/sec)
Estimated mixed
layer turnover rate
(d-t)
R.tt
(b)
0.95
4.2
1.29
~9.9x
I~
~ 1.9
0.74
13.4
0.89
~4.6x
I~
~0.9
0.97
2.5
~7.7x
I~
~9.'
0.14
60.9
0.21
153.6
0.91
7.9
0.20
37.1
0.82
14.1
0.19
36.3
0.07
88.S
0.36
~ 5.5 x 10~ 3.8 x 10-
~0.1
1.06
~ 1.1 x 10-1
~1.9
0.71
~ 1.2 x 1~
~ 5.9 x 10-
0.22
225
~3.2x
~2.6
~ 1.0
I~
x 10.•..•
" - time separation calculated for populations between the sudace and 1% light depth.
f - average chl/P ••• ratio (mole/mole) at the surface.
tt - ratio of PSU sizes between sudace and 1 % light depths.
ttt - ratio of P masJl values between surface and 1% light depths.
phyll field that is dependent upon turbulence can provide information about the
mixing history of a water column. Here I will examine the possibility that the size
of PSUs is one attribute that can be used to gain an understanding of the mixing
history (ie., chl/Proo ratios).
Consider the effects of turbulence on the light history of phytoplankton on GB
and in the NYB (e.g., Example 3). H the vertically uniform distribution of chlorophyll and phytoplankton species composition on GB throughout the year is a consequence of continuous tidal stirring, it can be hypothesized that the mixing processes
occur on a time scale that is as fast as or faster than that to which cells can adapt
to variations in the light regimes influenced by turbulence. This hypothesis is supported by the evidence that chl/P 700 ratios at the surface and 1% light depths
usually differ by less than 5% between the two isolumes (Table 2). In the midshelf
of the NYB in the winter (Fig. 4b) the chlorophyll distribution suggests that phytoplankton sinking rates exceed the rate of vertical mixing, so that a cell at the surface
might become progressively more shade adapted as it sinks deeper in the water
226
[41, 2
lournal of Marine Research
_
2000
....e
e
o
~
1800
1600
a..
~
••
1400
.c
u
1200
o
6
12
18
24
30
TIME (hrl
36
42
48
Figure S. Time-course study of the effect of nutrient enrichment on the size of photosynthetic
units in a natural phytoplankton community incubated at S % natural sunlight. The control
(.) containing less than 1.5 #tM total DIN and less than 0.5 #tM PO:- was incubated on
deck at sea-surface temperatures; after 6 h, NO. and PO:- (~) were added to concentrations
of 50 p.M and 10 p.M, respectively, and chl/P700 ratios were monitored.
column. This hypothesis is further supported by the evidence that chl/P 700 ratios at
the surface and 1% light depths differ by fivefold.
In this qualitative way it is suggested that vertical mixing processes influence not
only the distribution of the phytoplankton biomass but also the distribution of some
physiological, or biochemical, properties of the cells. To estimate the rate of vertical
displacement of phytoplankton in the water column from a physiological or biochemical attribute, a number of criteria must be met. First, the attribute must vary
on some time scale that is compatible with an understanding of vertical mixing
processes. Second, it must vary over such a dynamic range as to be applicable in a
study of the upper ocean. Third, it should not be influenced by a material resource
factor such as nutrient limitation. Finally, the attribute should not be limited by a
cropping factor such as grazing.
Time scale of change in chi/ProDratios. For chI/P700 ratios to be useful in understanding vertical mixing processes in the upper ocean, the time scale of change must
be compatible with mixing processes and, ideally, insensitive to first-order variations
in light (i.e., diel changes). In August 1977 a subsurface chlorophyll maximum was
observed in the midshelf of the New York Bight (Fig. 4d). The phytoplankton in
this layer were composed almost entirely of large Coscinodiscus spp., especially C.
concinnus (83%) and C. centralis (16%). A 36-h time-series station, made at
39°46.0' Nand 073°46.0' W, provided an opportunity to follow diel changes in
chl/P7oo ratios in situ.
The chI/P 700 ratios are plotted as a function of time of day in Figure 6. These
data suggest that PSU sizes did not systematically vary with diurnal variation in
incident light in situ in the subsurface chlorophyll maximum. However, the possi-
1983]
Falkowski:
e.•... 4000
••
•
E
• •
;;3000
a
•...
0..
.•...
2000
Adaptation
227
and mixing
• • • •
•
•
••
\)
&:.
Co)
1000
o
0400 0800 1200 160020002400
TIME (hr)
Figure 6. Effects of day-night variations on chlorophyll/P7\lO ratios in situ in a natural
plankton community consisting overwhelmingly of Coscinodiscus spp.
phyto-
bility that PSU sizes undergo diel periodicity cannot be completely ruled out. Owens
et al. (1980) have presented evidence suggesting that there is a diel periodicity on
C/chl ratios in situ in near-surface waters during a diatom bloom on the Long
Island Shelf. They inferred that this periodicity was primarily influenced by variations in intracellular chlorophyll. As it has been suggested that variations in chI/cell
in diatoms are primarily due to variations in chl/P700 ratios (Falkowski and Owens,
1980; Perry et al., 1981), diel periodicities in PSU sizes may occur near the surface
but not at low light depths. To avoid temporal anomalies in vertical profiles of PSU
sizes, samples were obtained at all depths simultaneously between 1000 and 1400
hours local time, and processed as rapidly as possible.
The rate constants for light-shade adaptation can be estimated from time-course
studies of changes in size of PSUs following the transfer of shade-adapted phytoplankton to high light intensities or of light-adapted phytoplankton to shade. During
the August 1977 Coscinodiscus bloom, a large phytoplankton sample was obtained
from the 1% light depth, and subsamples of these populations were incubated on
deck in natural sunlight at sea-surface temperatures. Chl/P700 ratios were frequently
measured over the next 48 h. These results are plotted as the function R, which is
the ratio of chl/P 700 ratios at times To and Tn, versus time (Fig. 7a).
Time-course studies of light-shade adaptation suggest that changes in chI/P7oo
ratios can be approximated by first-order kinetics as a function of changes in light
intensity (e.g., Falkowski, 1980; Rivkin et al., 1982). If the initial chl/P700 ratio R"
is normalized to 1.0, the rate of change of chl/P700 ratios can be approximated by
the equation:
228
lournal of Marine Research
[4t:;·2
I.
0.8
~
•.•
0
•.•
g 2
Q.
•..•.
:c
0.6
Q.
•..•.
A:
~0.4
•
0.2
4
8
12
16
TIME (hours)
20
24
48
o
...----.
9
"9
,x
0::
•
-I
•
.
0::, 0::
•
0
'--'
c
-2
.J
o
I
I
I
••
8
12
I
I
16
20
TIME (hours)
Figure 7. <a) Time course of changes in chl/P'IOOratios following light adaptation of a shadeadapted natural phytoplankton community, dominated by Coscinodiscus spp., in the New
York Bight. (b) The same data replotted in the log form, suggesting that light-shade adaptation follows first-order kinetics.
dR = -k(R -R ) e-kt
__ t
dt
0
""
,
(1)
where, k1 is a first-order rate constant.
Integration of Eq. (1) yields
(Rt-Roo)]
-kIt = In [ (Ro-Roo)
,
where Rt is the Chl/P700ratio at time t, Ro is the chl/P700 ratio at time zero, and
is the chl/P700 ratio at infinite time after adaptation.
(2)
RIIO
1983]
Falkowski: Adaptation and mixing
229
The data in Figure 7a are plotted in the integrated form in Figure 7b to estimate
k1• From these data, k1 is estimated as 5.5 x 10-2 h-1 at 25°e (r2
0.90, n = 11).
Similar rate constants for light-shade adaptation were calculated from changes in
chI/cell for individual species from batch and turbidostat culture experiments, and
these calculations are summarized in Table 3. In culture, changes in chI/ceD are
highIy correlated with changes in PSUs (Falkowski and Owens, 1980; Perry et al.,
1981). Laboratory data suggest that there is very little hysteresis between a light-toshade transition and a shade-to-light transition (Falkowski, 1983), but it is not clear
to what extent k1 is itself a function of the magnitude of change in the light field.
The data in Table 3 suggest that this effect may be small, but to avoid unnecessarily
complicating the field data, only the surface and 1% light depths were analyzed.
The effect of temperature on the rate constants for light-shade adaptation is somewhat uncertain in natural systems. Owens et aI. (1978) estimated a QI0 for chlorophyll biosynthesis in the neritic diatom, Skeletonema costatum, as ca 2.5. The data
of Hitchcock (1977) suggest a mean QIOin S. costatum and Detonuma confervacea
of ca 1.5 for changes in cellular chlorophyll. The discrepancies between the various
QIo.values may indicate that the temperature coefficient of light-shade adaptation
is species specific, and/or may reflect differences in sampling frequency by various
researchers. Assuming, however, that the rate of change of chljP700 ratios is dependent on the rate of enzymatic activity involved in chlorophyll metabolism, a
generalized QI0 of ca 2.0 is probably applicable in natural phytoplankton communities. If a QI0 of 2.0 is applied to the first-order rate constants calculated for
light-shade adaptation given in Table 3, the average for ki at 15°e is estimated 88
(2.8 ± 0.9) X 10-2 h-1•
Using this value of k1, it is possible to calculate how far apart in time samples
taken from depths z and z ± /J.z are by rearranging Eq. (2) and solving for t, such
that
=
t
= _1_
-k1
In [(Rt-R..,)].
(3)
(Ro-R..,)
The effect of temperature modification on ki can be modeled by the equation:
k1 = 0.009geo.•o.G8T
,
(4)
where T is in °e.
Dynamic range. To solve Eq. (2) numerically, the dynamic range (Ro-R~) must
be known. If we take the initial high light-adapted chI/P7o.0 ratio (Ro) of a given
population to be 1.0, the dynamic range will be given by the maximum chI/P7o.o.
ratio measured in a shade-adapted population.
The primary environmental factor influencing chI/P 70.0.ratios is light. Over a
range of light intensities from 2000 to 20 p.E m-2 sec-I (ca the range from 100 to
1% of full sunlight), the change in ChI/P7o.o.
ratios is maximal (Falkowski and Owens,
[41, 2
lournal of Marine Research
230
.•....
•...
00
Cl\
.....
"-'
*
fc
~
b....•
b....•
b....•
x
X
\0
.....
N
N
x
•...
N
X
1
1
o
.....
....• o
x x
\0
t-;
b....•
b....•
.....
N
b....•
x
X
00
N
N
M
b....•
X
N
q
•...
M
d
N
.•..•
o
c1
'"....•
o
N
o
.....
o
I':
..•..
<)
u
<)
o
SO
....l
N
....•
....••
N
.....
o
N
'"N
oS
1983]
Falkowski: Adaptation and mixing
231
1980). Below 20 JiB m-2 sec-I, cells tend to bleach, and there is a concomitant
decrease in chl/P 100ratios. Over the range of light intensities a cell would experience
in the euphotic zone, the ratio of chl/P 100obeys a log function of light intensity
(Falkowski and Owens, 1980) and therefore would be expected to vary linearly
with depth in the water column (Falkowski and Wirick, 1981). A minimum value
for the ratio of chl/P100 values for fully light-adapted and fully shade-adapted cells,
both in culture (see Falkowski and Owens, 1980; Perry et aI., 1981; Ley and
Mauzerall, 1982) and in the field, is ca 0.067; i.e., chl/Pm ratios at the 1% light
depth do not appear to be more than fifteen times the' chl/P100 ratio at the 100%
light depth (Table 2). The value of R«J in Eq. (2) is taken to be 0.067, and thus
(Ro-R«J) = 0.933.
Resource factors. Of the limiting resource factors potentially influencing the distribution of chl/P100 ratios, inorganic nitrogen and/or phosphate are likely to have
the greatest infiuence. In the summer, in the upper mixed layer of the NYB, inor:ganic nitrogen is frequently undetectable; however, phytoplankton assimilation numbers are often maximal (Falkowski, 1981). In August 1977, whole natural seawater
samples were taken in the midshelf of the NYB at 12 m, which corresponded to
ca 5% light depth. These samples, containing less than 1.5 p,M total dissolved inorganic nitrogen (N03- + N02- + NH4+) and less than 0.50 p,M P042-, were
enriched to 50 p,M with NOa- and to 10 p,M with P042-. Over the next 72 h, the
samples were incubated on the deck in 20-1 polypropylene carboys at sea-surface
temperatures. The carboys were screened to 5 % light and incubated in natural
sunlight while ch1jP100ratios were monitored.
In this experiment (Fig. 5) the Chl/P 700ratios did not change significantly in
nutrient enriched samples, as compared with unenriched samples, during the first
24 h, and increased thereafter. These results suggest that intracellular reserves of
nutrients may be adequate to maintain the integrity of the size of PSUs for about
a day.
Nutrient limitation may decrease the size of PSUs by forcing chlorosis (Falkowski,
1980). Off the coast of Hawaii, Chl/P700ratios are lower than those generally observed under similar temperature and nutrient regimes in the NYB. These results
suggest that chl/P 100ratios may be influenced by nutrient limitation in oligotrophic
regions.
Cropping factors. There are two major cropping factors infiuencing the distribution of phytoplankton: grazing and sinking. The ratio of the number of chlorophyll
molecules to the number of P700molecules is an intrinsic property of a given cell.
Unlike phaeophytin/cWorophyll ratios, grazing stress is unlikely to alter chlorophyll/P700 ratios. While the vertical distribution of chlorophyll might be in1luenced
by vertical variations in grazing pressure, the distribution of chl/P700 ratios within
232
lournal of Marine Research
[41, 2
the chlorophyll profile will not be affected; the chemical composition of the ungrazed
cells is not inftuenced by those cells which have already been consumed.
Sinking rates are likely to affect the vertical distribution of chl/P 700 ratios in the
water column, as these ratios are inftuenced by changes in light intensity. As the
vertical displacement of cells in a turbulent ocean is a balance between sinking and
mixing, changes in chl/P700 ratios potentially can be used to estimate the net displacement rate at phytoplankton in situ.
Estimating vertical mixing rates. The data and arguments presented in the pre~ng
sections suggest that chl/P700 ratios can fit the criteria required for estimating
vertical mixing rates of phytoplankton in natural waters. If it is accepted that the
effects of turbulence can inftuence the light regime experienced by a cell, and that
a.cell can adapt to variations in light regimes by altering the size of photosynthetic
units (i.e., it can light-shade adapt via changes in chl/P700 ratios), it is possible to
estimate the rate at which phytoplankton are translocated by vertical mixing processes"from knowledge of the vertical variations in chl/P700 ratios.
By modifying Eq. (3), the rate at which phytoplankton at depth z are displaced
to depth Z ± Az is given by the equation:
Az _
At -
Az(-k,)
In(Rt-R",)
(Ro-R",)
(5)
Using Eq. (5), apparent maximum and/or minimum vertical displacement rates
are estimated for various cases on GB, in the NYB, and in the SP (Table 2). These
calculations are average displacement rates or piston velocities for the entire euphotic
zone (i.e., to the 1% light depth) and include the temperature correction inferred
from Eq. (4).
The results of these calculations suggest that vertical displacement rates vary by
about two orders of magnitude in the regions studied (from ca 10-1 to 10-8 em/sec).
If the mean vertical displacement rates in the euphotic zone (Z6) are applied to the
upper mixed layer (Z•••
), it is possible to estimate turnover times for mixed layer
whenZ", > Ze. These estimates range from ca 0.1 d-1 to 10 d-1•
Without direct simultaneous measurements of vertical mixing, it is difficult to
verify these calculations with physical evidence. The least temporal variance in
the estimated vertical displacement rates is observed on GB. The data (Table 2)
suggest that mixing rates in this region are only about two times greater in the early
spring than in the early fall. Although the region is continuously mixed by tidal
stirring, the potential energy (positive buoyancy) of the water column increases as
the water column heat content increases. In the fall, the water column on GB requires more kinetic energy to dissipate vertical heat gradients. As tidal energy available for mixing does not increase in warmer months, the rate at which density
gradients are dissipated is decreased. The increased vertical stability on GB in the
1983]
233
Falkowski: Adaptation and mixing
early fall, associated with increased heat content, is physiologically reflected in the
tendency of phytoplankton to become more shade adapted near the bottom of the
euphotic zone; i.e., their apparent residence time at the bottom of the euphotic zone
is increased. Thus, the ratio of PSU sizes between the surface and 1 % light depths
decreases from 0.95 in the spring to 0.78 in the fall. The vertical displacement rates
on GB estimated by the PSU method described here agree well with estimations of
vertical mixing based on heat flux calculations (Hopkins and Garfield, 1981).
Compared with GB, the vertical mixing rates in the NYB appear to be much
more variable (a factor of -30). On GB tidal stirring is a very predictable source
of energy for mixing, while in the NYB the major energy sources for mixing are
wind stress and convective overturn, which are much more stochastic forcing functions. It seems counter-intuitive that vertical mixing rates in the middle of March
in the middle of the continental shelf of the NYB could be lower than in subtropical
Pacific near Hawaii (Table 2). The results of the calculations presented here do not
necessarily reflect the "average" vertical mixing condition for the season and region
indicated. For example, on March 15, 1977, the day the midshel£ station in the
NYB was sampled, wind stress was very low (-0.1 dyne/em2) and only a small
portion of this wind energy is used for vertical mixing (Kullenberg, 1976). This
relatively low wind stress, in combination with the early onset of thermal stratification (Fig. 4b), apparently resulted in low vertical mixing. Qualitatively low mixing
rates could be inferred from the vertical chlorophyll profile. The phytoplankton
were dominated by net plankton (> 20 fLm) diatoms, which have sinking rates on
the order of 10-8 em/sec (Smayda and Boleyn, 1966; Bienfang, 1980). Vertical
mixing rates greatly in excess of this sinking rate would tend to equalize the distribution of chlorophyll, while mixing rates equal to or less than this sinking rate would
lead to an increase of chlorophyll near the bottom of the water column (Bienfang,
1980).
Of the vertical mixing calculations presented, probably the most unreliable is that
for the SP. Because of extreme nutrient impoverishment at the surface and not at
the 1% light depth, the dynamic range of the P100 technique was approached, such
that the calculated mixing rate has to be taken as a maximum. It is highly likely that
the actual mixing rates in this region were lower.
It is interesting to note that there is apparently a linear relationship between Rt
and the ratio of PmaxB (in fLg C fLg chl-1 h-1) values obtained at the surface and
1% light depths from P vs 1 curves. Based on this small data set (n
6), the correlation coefficient between Rt and the ratio of PmaxB values is 0.91 and [Rt = 0.97
=
~~max:~z" 0.25]. These results suggest that the ratio of light-saturated assimilation
max
Ze
numbers simultaneously obtained from the surface and 1% light depths in shortterm incubations could be used to estimate vertical mixing in a fashion similar to
that described for chl/P7oo ratios.
234
Journal of Marine Research
[41, 2
Vertical mixing plays an important role in the regulation of primary production
in each of the three regions examined. On GB, primary production is estimated as
ca 450-500 g C m2 yr-l (Cohen et al., 1981; Falkowski, unpubl). On the shallow
southeast portion of the Banks the critical depth almost always exceeds the bottom,
and tidal stirring, along with the normal circulation around the region (Hopkins and
Garfield, 1981), provides sufficient nutrient inputs to drive the high production. In
the NYB, primary production is estimated at 250-300 g C m2 yr-l, 40% of which
occurs in the spring bloom (March to April). In the summer, strong vertical density
gradients in the NYB prevent the vertical mixing of nutrient-rich water into the
euphotic zone, which in tum leads to a reduced annual production compared with
nearby GB. If, during the most productive times of the year, the water column is
not mixed at a rate exceeding phytoplankton sinking rates, there will be a tendency
for poorer coupling between primary and secondary pelagic production (Dagg and
Turner, 1983). The less predictable mixing regime in the NYB may result in
relatively large losses of phytoplankton due to sinking and subsequent export
alsh
rw
et al., 1981).
4. Conclusions
In the early years of oceanography, phytoplankton ecologists used floristic evidence and their taxonomic skills to trace water masses. The data and arguments
presented here suggest that, on a smaller scale, physiological features of phytoplankton can be used as means of understanding turbulence in the upper ocean. While the
specific technique described here is hardly routine, in future years other simpler
methods, such as fluorescence induction (Harris, 1980; Fujita, per. comm.) , which
also use light gradients and physiological responses, may be used more routinely to
estimate vertical mixing.
Acknowledgments.
I thank R. W. Eppley, T. S. Hopkins, A. Okubo, T. C. Malone, and C.
Wirick for valuable discussions and criticisms, and T. C. Owens and P. Bienfang for assistance
in data collection. I am grateful to J. J. Walsh for his continued support.
This research was performed under the auspices of the United States Department of Energy
under Contract No. DE-AC02-76CHOO016.
By acceptance of this article, the publisher and/or recipient acknowledges the U.S. Government's right to retain a nonexclusive, royalty-free license in and to any copyright covering this
paper.
REFERENCES
BeardalI, J. and I. Morris. 1976. The concept of light intensity adaptation in marine phyto., plankton: some experiments with Phaeodactylum tricornutum. Mar. Bio., 37, 377-·387.
Beardsley, R. C., W. C. Boicourt and D. V. Hansen. 1976. Physical oceanography of the Middle
Atlantic Bight, in Middle Atlantic Continental Shelf and the New York Bight, M. G. Gross,
. ed., Limnol. Oceanogr. Spec. Symp. No.2, 20-34.
1983]
235
Falkowski: Adaptation and mixing
Bienfang, P. K. 1980. Phytoplankton sinking rates in oligotrophic waters off Hawaii, U.S.A.
Mar. Bio!', 61, 69-77.
Bigelow, H. B. 1933. Studies on the waters on the continental shelf, Cape Cod to Chesapeake
Bay. I. The cycle of temperature. Pap. Phys. Oceanogr., 2, 135 pp.
Bumpus, D. F. 1973. A description of the circulation on the continental shelf of the east coast
of the United States. Prog. Oceanogr., 6,111-157.
Cohen, E. B., M. D. Grosslein, M. P. Sissenwine, F. Steimle and W. R. Wright. 1981. An energy budget of Georges Bank, in Multispecies Approaches to Fisheries Management. Can. J.
Fish. Aquat. Sci. Spec. Symp. (in press).
Cullen, J. J. and R. W. Eppley. 1981. Chlorophyll ma.~mum layers of the Southern California
Bight and possible mechanisms of their formation and maintenance. Oceanol. Acta, 4, 23-32.
Dagg, M. J. and J. T. Turner. 1983. The impact of zooplankton grazing in waters of intermediate
and high primary production, on the continental shelf of the Northeastern United States. J.
Plankton Res., (submitted).
Eppley, R. W. and J. E. Brooks. Residence time at depth of phytoplankton communities: inferences from photosynthesis characteristics, (unpublished manuscript).
Eppley, R. W., E. H. Renger and W. G. Harrison. 1979. Nitrate and phytoplankton production
in southern California coastal waters. limnoI. Oceanogr., 24, 483-494.
Falkowski, P. G. 1980. Light-shade adaptation in marine phytoplankton, in Primary Productivity
in the Sea, P. G. Falkowski, ed., Plenum Press, New York, 531 pp.
-1981. light-shade adaptation and assimilation numbers. J. Plankton Res., 3, 203-216.
-1983. Kinetics of light intensity adaptation in Dunalie/la tertiolecta, a marine plankton
chIorophyte. Photosynthetica, (in press).
Falkowski, P. G., T. S. Hopkins and J. J. Walsh. 1980. An analysis of factors affecting oxygen
depletion in the New York Bight. J. Mar. Res., 38, 479-506.
Falkowski, P. G. and T. G. Owens. 1980. light-shade
adaptation: two strategies in marine
phytoplankton. Plant Physio!., 66, 592-595.
Falkowski, P. G., T. G. Owens, A. C. Ley and D. MauzeraII. 1981. Effects of growth irradiance
levels on the ratio of reaction centers in two species of marine phytoplankton. Plant Physio!.,
68, 969-973.
Falkowski, P. G., J. Vidal, T. S. Hopkins, T. E. Whitledge and W. G. Harrison. 1983. Summer
nutrient dynamics in the Middle Atlantic Bight: Primary production and utilization of phytoplankton carbon. J. Plankton Res., (submitted).
Falkowski, P. G. and K. von Bock. 1979. Atlantic Coastal Experiment. Phytoplankton species
composition: 1976-1978. BNL Data Report 26335.
Falkowski, P. G. and C. D. Wirick. 1981. A simulation model of the effects of vertical mixing
on primary productivity. Mar. BioI., 65, 69-75.
Harris, G. P. 1980. The relationship between chlorophyll a fluorescence, diffuse attenuation
changes and photosynthesis in natural phytoplankton populations. J. Plankton Res., 2, 109127.
Herron, H. A. and D. MauzeralI. 1971. The development of photosynthesis in a greening mutant of Chlorella and an analysis of the light saturation curve. Plant PhysioI., 50, 141-148.
Hitchcock, G. L. 1977. The time course of photosynthetic adaptation, the growth rate response,
and variations in the pigment, carbohydrate, and protein content of Skeletonema costatum
and Detonula confervacea to changes in light intensity. Ph.D. thesis, Univ. Rhode Island.
Hiyama, T. and B. Ke. 1972. Difference
Biophys. Acta, 267, 160-171.
spectra and extinction
coefficients of
P'llIO.
Biochim,
236
lournal of Marine Research
[41, 2
Hopkins, T. S. and N. Garfield. 1981. Physical origins of Georges Bank water. I. Mar. Res.,
39, 465-500.
IS'Srgensen, E. G. 1969. The adaptation of plankton algae IV. Light adaptation in different algal
species. Physiol. Plant., 22, 1307-1315.
Ketchum, B. H. and N. Corwin. 1964. The persistence of "winter" water on the continental
shelf south of Long Island, New York. Limno1. Oceanogr., 9, 467-475.
King, F. D. and A. H. Devol. 1979. Estimates of vertical eddy diffusion through the thermocline from phytoplankton
uptake rates in the mixed layer of the eastern tropical Pacific.
Limno1. Oceanogr., 24, 645-651.
Kullenberg, G. E. B. 1976. On vertical mixing and the energy transfer from the wind to the
water. Tellus, 28, 159-165.
Ley, A. C. and D. Mauzerall. 1982. Absolute absorption cross-sections for photosystem II and
. the minimum quantum requirement for photosynthesis in Chiorella vulgaris. Biochem. Biophys. Acta, 680, 95-106.
Malone, T. C. 1971. Diurnal rhythms in net plankton and nannoplankton assimilation ratios.
Mar. BioI., 10,285-289.
-; 1977. Light-saturated photosynthesis by phytoplankton size fractions in the New York
Bight, U.S.A. Mar. BioI., 42,281-292.
Malone, T. C., P. G. Falkowski, T. S. Hopkins, G. T. Rowe and T. E. Whitledge. 1983: Mesoscale response of diatom populations to a wind event in the plume of the Hudson River. Deep
Sea Res., (in press).
Marra, I. 1978. Effect of short-term variations in light intensity on photosynthesis of a marine
phytoplankter: a laboratory simulation study. Mar. BioI., 46, 191-202.
Myers, I. and I. Graham. 1971. The photosynthetic unit in Chlorella measured by repetitive
short flashes. Plant Physio1., 48, 282-286.
Owens, T. G., P. G. Falkowski and T. E. Whitledge. 1980. Diel periodicity in cellular chlorophyll content in marine diatoms. Mar. BioI., 59, 71-77.
Owens, T. G., D. M. Riper and P. G. Falkowski. 1978. Studies of S-aminolevulinic acid de. hydrase from Skeletonema costatum, a marine plankton diatom. Plant Physiol., 62, 516-521.
Perry, M. I., M. C. Talbot and R. S. Alberte. 1981. Photoadaptation
in marine phytoplankton:
response of the photosynthetic unit. Mar. BioI., 62, 91-101.
Prezelin, B. B. and B. M. Sweeney. 1979. Photoadaptation
of photosynthesis in two bloomforming dinoflagellates, in The Second International
Conference on Toxic Dinoflagellate
Blooms, D. L. Taylor and H. H. Selinger, eds., Elsevier, New York, 101-106.
Reid, F. M. H., E. Stewart, R. W. Eppley and D. Goodman. 1978. Spatial distribution of phytoplankton species in chlorophyll maximum layers off Southern California. Limno1. Oceanogr.,
23, 219-226.
Riley, G. A. 1965. A mathematical model of regional variations in plankton. Limnol. Oceanogr.,
10, R202-215.
Riper, D. M., T. G. Owens and P. G. Falkowski. 1979. Chlorophyll turnover in Skeletonema
costatum, a marine plankton diatom. Plant Physio1., 64, 49-54.
Rivkin, R. B., H. H. Seliger, E. Swift and W. H. Biggley. 1982. Light-shade adaptation by the
oceanic dinoflagellates, Pyxocystis rochiluca and P. fusiformis. Mar. BioI., 68, 181-191.
Ryther, I. H. and E. M. Hulburt. 1960. On winter mixing and the vertical distribution of phytoplankton. Limnol. Oceanogr., 5, 337-338.
Ryther, J. H. and D. W. Menzel. 1959. Light adaptation by marine phytoplankton. Linmol.
Oceanogr.,4, 492-497.
Shibata, K. 1958. Spectrophotometry of intact biological materials. I. Biochem., 45, 599-623.
1983]
Falkowski: Adaptation and mixing
237
Shiozawa, I. A., R. S. Alberte and I. P. Thornber. 1974. The P...,-chlorophyll a-protein: Isolation and some characteristics in higher plants. Arch. Biochem. Biophys., 165, 388-397.
Smayda, T. I. and B. I. Boleyn. 1966. Experimental observations on the flotation of marine
diatoms. II. Skeletonema costatum and Rhizosolenia setigera. Limnol. Oceanogr., 11, 18-33.
Steemann Nielsen, E. and V. Kr. Hansen. 1959. Light adaptation in marine phytoplankton
population and its interrelation with temperature. Physio!. Plant., 12, 353-370.
-1961. Influence of surface illumination on plankton photosynthesis in Danish waters (56N)
throughout the year. Physio!. Plant., 14, 595-613.
Steemann Nielsen, E. and T. S. Park. 1964. On the time course in adapting to low light intensities in marine phytoplankton. I. Cons. Int. Explor. Mer., 29, 19-24.
Strickland, I. D. H. and T. R. Parsons. 1972. A Practical Handbook of Seawater Analysis. Fish.
Res. Bd. Can. Bull., 167, 2nd edition, Ottawa, 310 pp.
Sverdrup, H. U. 1953. On conditions for the vernal blooming of phytoplankton.
I. Cons. Int.
Explor. Mer., 18, 287-295.
Thornber, P. I., R. S. Alberte, F. A. Hunter, I. A. Shiozawa and K.-S. Kan. 1977. The organization of chlorophyll in the plant photosynthetic unit. Brookhaven Symposium Bio!., 28,
132-148.
Walsh, I. I., G. T. Rowe, R. L. Iverson and C. P. McRoy. 1981. Biological export of shelf
carbon: a neglected sink of the global CO. cycle. Nature, 291, 196-201.
Whitledge, T. E., S. C. Malloy, C. I. Patton and C. D. Wirick. 1981. Automated nutrient
analyses in seawater. BNL 51398.
Yentsch, C. S. and D. W. Menzel. 1963. A method for the determination of phytoplankton,
chlorophyll, and phaeophytin by fluorescence. Deep-Sea Res., 10,221-231.
Received: 15 April, 1981; revised: 17 January, 1983.