1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
YECSS4766_proof ■ 16 May 2015 ■ 1/13
Estuarine, Coastal and Shelf Science xxx (2015) 1e13
Contents lists available at ScienceDirect
Estuarine, Coastal and Shelf Science
journal homepage: www.elsevier.com/locate/ecss
Q6
Q5
Cell volumes of marine phytoplankton from globally distributed
coastal data sets
P.J. Harrison a, *, Adriana Zingone b, M.J. Mickelson c, S. Lehtinen d, N. Ramaiah e,
A. Kraberg f, J. Sun g, A. McQuatters-Gollop h, H.H. Jakobsen i
a
Dept. Earth & Ocean Sciences, University of British Columbia, Vancouver, BC V6T 1Z4, Canada
Stazione Zoologia Anton Dohrn, Villa Comunale, Italy
Massachusetts Water Resources Authority, Charlestown, MA 02129, USA
d
Marine Research Center, Finnish Environmental Institute (SYKE), 00251 Helsinki, Finland
e
CSIR-National Institute of Oceanography, Goa, 403 004 India
f
Biologische Ansatalt Heloland, Alfred Wegener Institute Helmhotz Centre for Polar and Marine Research, 27498 Helgoland, Germany
g
College of Marine Science & Engineering, University of Science & Technology, Tianjin, 300457, PR China
h
Sir Alister Hardy Foundation for Ocean Science, Plymouth, PL1 2PB, UK
i
Bioscience, Roskilde 4000, Aarhus University, Denmark
b
c
a r t i c l e i n f o
a b s t r a c t
Article history:
Accepted 4 May 2015
Available online xxx
Globally there are numerous long-term time series measuring phytoplanton abundance. With appropriate conversion factors, numerical species abundance can be expressed as biovolume and then converted to phytoplankton carbon. To-date there has been no attempt to analyze globally distributed
phytoplankton data sets to determine the most appropriate species-specific mean cell volume. We have
determined phytoplankton cell volumes for 214 of the most common species found in globally distributed coastal time series. The cell volume, carbon/cell and cell density of large diatoms is 20,000, 20,000
and 0.1 times respectively, compared to small diatoms. The cell volume, carbon/cell and cell density of
large dinoflagellates is 1500, 1000 and 0.7 times respectively, compared to small dinoflagellates. The
range in diatom biovolumes is > 10 times greater than across dinoflagellates (i.e. >20,000 vs. 1500 times)
and within any diatom species, the range in biovolume is up to 10-fold. Variation in diatom cell volumes
are the single largest source of uncertainty in community phytoplankton carbon estimates and greatly
exceeds the uncertainty associated with the different volume to carbon estimates. Small diatoms have 10
times more carbon density than large diatoms and small dinoflagellates have 1.5 times more carbon
density than large cells. However, carbon density varies relatively little compared to biovolume. We
recommend that monthly biovolumes should be determined on field samples, at least for the most
important species in each study area, since these measurements will incorporate the effects of variations
in light, temperature, nutrients and life cycles. Since biovolumes of diatoms are particularly variable, the
use of size classes will help to capture the percentage of large and small cells for each species at certain
times of the year. This summary of global datasets of phytoplankton biovolumes is useful in order to
evaluate where locally determined biovolumes lie within the global spectrum of spatial and temporal
variations and may be used as a species cell volume reference where no locally determined volume
estimates are available. There is a need to adopt standard protocols for estimating biovolumes and
documenting the accompanying metadata which would improve inter-comparability among time series
data sets.
© 2015 Published by Elsevier Ltd.
Regional index terms:
Global
Keywords:
phytoplankton cell volume
cell carbon
diatoms
dinoflagellates
carbon biomass
biovolume
1. Introduction
* Corresponding author.
E-mail address: [email protected] (P.J. Harrison).
There is considerable concern about the long-term changes that
are occurring in coastal ecosystems, leading to the development of
management strategies and mitigation procedures to deal with
current and future anthropogenic stressors and climatic changes.
http://dx.doi.org/10.1016/j.ecss.2015.05.026
0272-7714/© 2015 Published by Elsevier Ltd.
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56 Q2
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 2/13
2
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
One way to capture this long-term variability in phytoplankton is to
set up time series sampling stations and document the variations in
abundance and species composition in relation to changes of
environmental variables (Zingone et al., 2010). Two of the most
important variables in phytoplankton time series are the estimates
of the carbon biomass as a common currency, and the abundance of
different species since they shape the planktonic food web and
determine the productivity of the whole pelagic ecosystem. Chl a as
a proxy for determining phytoplankton carbon is frequently used,
but there are large variations in the C/Chl a ratio among and within
a species due to environmental factors such as seasonal changes in
temperature and limitations in nutrients and light (Taylor et al.,
1997). More importantly, when bulk measures such as Chl a are
used, no information is obtained on the amount of carbon that is
contributed by individual species.
For various ecosystem applications and modeling, it is necessary
to convert phytoplankton cell counts into a common currency such
as wet weight, carbon or nitrogen, because a large number of small
cells are equivalent to a few very large cells in terms of carbon
biomass that is utilized as food for the next trophic level. To convert
cell numbers to carbon biomass for primary producers, it is
necessary to know the cell volume of the various species in the
sample, and the carbon per cell volume (carbon density) multipliers
for each species (Verity et al., 1992; Montagnes et al., 1994;
Menden-Deuer and Lessard, 2000).
There are several ways to calculate cell volumes. The ‘gold standard’ is to determine the geometrical shape that approximates the
shape of the cell and then make measurements of the dimensions to
enter into the formula for that particular geometrical shape (Mullin
et al., 1966; Strathmann, 1967; Eppley et al., 1970; Taguchi, 1976;
Wheeler, 1999). Some of the challenges in this approach are that
different investigators may choose a different geometric shape than
the recommended shape (Hillebrand et al., 1999; Sun and Liu, 2003)
for the same species, especially for cells with a complex shape. For
small cells, it may be difficult to measure the dimensions accurately
due to the ‘halo effect’ around a small cell under the light microscope. In addition, the ‘hidden dimension’ (i.e. the depth dimension)
is difficult to measure since cells are viewed in two dimensions
under the microscope. Yet the use of microscopically determined
cell volumes is the only way to resolve C biomass estimates at the
species level (Montagnes et al., 1994; Menden-Deurer and Lessard,
2000). Alternatively, there are several automated or semiautomated methods for estimating cell volume include the Coulter
counter (Boyd and Johnson, 1995), image analysis (FlowCAM)
(Sieracki et al., 1998; Jakobsen and Carstensen, 2011), and flow
cytometry (Olson et al., 1985), although they all have some limitations. A recent improvement is the direct measurement of a biovolume in 3D confocal microscopy, which however has been tested
for only a few selected species (Roselli et al., 2015).
Under the influence of seasonally varying environmental factors, the cell volume of diatoms often varies during the season with
different size classes occurring for a species (Olenina et al., 2006;
Jakobsen et al., 2015). When cells become nutrient (N, P or Fe)
limited, they are usually smaller (Harrison et al., 1990; Davidson
et al., 2002; Timmermans and van der Wagt, 2010; Edwards et al.,
2011). In contrast, diatoms become larger under silicate limitation
because there is not enough silicate for the two daughter cells to
complete the siliceous valves between them and therefore they
form a biprotoplastic cell (Harrison et al., 1977). Under light limitation, cells are usually smaller (Thompson et al., 1991). There is no
consistent trend with temperature since cell volume has been reported to decrease (Montagnes and Franklin, 2001) or increase with
increasing temperatures (Thompson et al., 1992). Under a range of
salinities from 5 to 25 in an Indian estuary, Mitra et al. (2012) found
that cells were smaller at higher salinities.
In addition to environmental factors, cell size varies during life
cycles. The importance of asexual and sexual reproduction in
diatom life cycles and the relation to variations in cell size is well
documented (von Dassow et al., 2006; D'Alelio et al., 2010). Sexual
reproduction can be induced by environmental stresses such as
nutrient limitation since sexual reproduction is more readily
inducible in small cells (Harrison et al., 1976; Costello and
Chisholm, 1981; Edlund and Stoermer, 1997; von Dassow et al.,
2006). One of the advantages of sexual reproduction for diatoms
is that the return to a large cell volume usually coincides with a
much higher growth rate (i.e. a reinvigoration or rejuvenation of
the cell's physiological processes) (Harrison et al., 1976; Costello
and Chisholm, 1981; Saravanan and Godhe, 2010) and surprisingly, a lower sinking rate for new larger post-auxospore cells of
Ditylum (Waite and Harrison, 1992). Sexual reproduction may occur
at various times, but at least in diatoms, there is a tendency for sex
to occur in the autumn when cells are smaller probably due to
summer nutrient limitation (Mizuno and Okuba, 1985; Waite and
Harrison, 1992; Koester et al., 2007; D'Alelio et al., 2010; von
Dassow and Montresor, 2011). An abrupt increase in cell volume
may also occur vegetatively and is termed vegetative enlargement
(Gallagher, 1983; Nagai et al., 1995).
Diatom cell sizes range from a few microns up to 2 mm (i.e.
1000x) and consequently their biovolumes can span about 9 orders
of magnitude. Therefore, it is necessary to be able to convert cell
abundance into cellular carbon, especially for trophic models.
Because determining biovolume microscopically is tedious and
time consuming, it is not surprising that there are only a few data
sets available, even for ecologically important species associated
with time series programs. Furthermore, few of these data have
ever been published.
Many time series have abundance data but they lack the
species-specific cell volume data to convert abundance into carbon
biomass. Hence, there is a need for a reference list of biovolumes for
a large number of species from different coastal sites. The objective
of this study was to collect and analyze biovolume data for the most
common (i.e. occur >5 times) phytoplankton species found in
globally distributed coastal time series data sets. Biovolumes have
been determined for Baltic Sea phytoplankton (Olenina et al., 2006)
and for some diatoms (Leblanc et al., 2012), but to-date there has
been no attempt to analyze global data sets to determine the
variation in cell volume for a large number of phytoplankton species from various coastal oceans.
2. Methods
2.1. The data sets
We obtained 40 published and unpublished cell volume data
sets from various coastal regions around the world that were often
produced in conjunction with time series monitoring programs.
The several datasets from northern San Francisco Bay and from
Chesapeake Bay were merged into one data set for each area to
avoid duplication, making a total of 35 sites (Table 1). The data sets
were cleaned up by merging synonyms, correcting spelling mistakes, removing non-relevant species and up-dating the nomenclature using WoRMS (World Registry of Marine Species; (https://
marinespecies.org)). This important process was by far the most
laborious step to harmonize and aggregate these 35 data sets into
one data base (Table 2). ALGAEBASE (http://www.algaebase.org/)
was used for additional nomenclatural validation and up-dating
and in some cases, more recent literature not yet incorporated
into either database was followed. These data will eventually be
placed in the SCOR Working Group 137 domain (wg127.net).
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 3/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
3
Table 1
Location of stations and the contributors of the biovolume datasets in Table 2. Chesapeake Bay is a combination of 4 data sets (Chesapeake Bay, Virginia, Maryland and Rhode
River) and northern San Francisco Bay combined 2 datasets (Collinsville and Suisun Bay).
Locality
North Atlantic, western
Narragansett Bay, Rhode Island, USA
Chesapeake Bay, USA
Neuse Estuary, North Carolina, USA
Shelf off NE Venezuela
North Atlantic, eastern
Baltic Sea, Gulf of Finland
Baltic Sea, Sweden
Baltic Sea, outer area, Denmark
Baltic Sea, Germany
North Sea, Germany
North Sea, southern, Germany
Mediterranean and Black Sea
Mediterranean, northwest, Balearic Sea
Mediterranean, Gulf of Naples, Italy
Mediterranean, Fusaro Lagoon, Italy
Mediterranean, Salento peninsula, SE Italy
Mediterranean, South of Cyprus
Aegean Sea, Lesvos, Greece
Adriatic Sea, southern, Mali Ston Bay, Crotatia
Black Sea, Turkey (2012)
Black Sea, Turkey (2014)
South Atlantic, western
Beagle Channel, Argentina/Chile
Straits of Magellan, Argentina/Chile
Indian Ocean
Arabian Sea, Karachi coast, Pakistan
Arabian Sea, northeast, Goa coast, India
Bay of Bengal, Indian Sundarbans (west), India
Bay of Bengal, Indian Sundarbans (east), India
Australia, west coast
North Pacific, western
Yellow Sea, eastern, Gyeonggi Bay, Korea
Bohai Sea, China
Yellow Sea, China
Yellow Sea, China
Yellow Sea, Jiaozhou Bay, China
North Pacific, eastern
Gulf of Alaska, USA
Coastal Waters of British Columbia, Canada
Dabob Bay, Washington, USA
Northern San Francisco Bay, USA
Southern San Francisco Bay, USA
Southern San Francisco Bay, USA
South Pacific, eastern
n Bay, Chile
Concepcio
Contributor
Smayda T, Graduate School Oceanography, Univ. Rhode Island, USA
Marshall HG, Lacoutre RV, Egerton T, Sellner KG, Hedrick S, Gallegos CL, Johnson JM
Paerl H, UNC, Institute of Marine Sciences, Chapel Hill NC, USA
Lugo-Vencaino et al. (2003)
Marine Research Centre, Finnish Environment Institute(1993e2013)
Swedish Meteorological Hydrological Inst. Gothenburg (1983e2012)
Aarhus University, Denmark (1984e2014)
Holstein (LLUR), Germany
€bel J, Jaschinski S, SAAERA, Germany
Go
Wiltshire KH, Kraberg A, Biologische Anstalt Helgoland (AWI) Germany
Sarno D, Zingone A, Stazione Zoologica Anton Dohrn, Naples, Italy
Sarno D, Zingone A, Stazione Zoologica Anton Dohrn, Naples, Italy
Sarno et al. (1993)
Stanca et al. (2013)
Tsagaraki T, Paraskevi P, Hellenic Centre for Marine Research, Crete, Greece cited in Tsagaraki et al. (2013)
Same as above e cited in Tsagaraki et al. (2013)–
Jasprica N, Caric M (1997)
Moncheva et al. (2012)
Moncheva et al. (2014)
Almandoz et al. (2011)
Zingone et al. (2011)
Naz et al. (2013)
Alkawri AAS, Ramaiah N, CSIR, National Inst. Oceanography, Goa, India
Mitra et al. (2012)
Mitra et al. (2012)
Thompson P, CSIRO Oceans & Atmosphere, Hobart, Australia
Jahan R, Choi JK, Dept. Oceanography, Inha University, Korea
Sun J. College of Marine Science and Engineering, Tianjin Univ., China
Sun J. College of Marine Science and Engineering, Tianjin Univ., China
Wang et al. (2011)
Sun et al. (2000)
Peterson T, Oregon Health & Science University, Portland OR, USA
Haigh R, Pacific Biological Stn, Nanaimo BC, Canada
Postel RJ, Horner R, Dept. Oceanography, Univ. Washington, USA
Lehman P, Dept. Water Resources, Div. Envir. Services, CA, USA
Lehman P, Dept. Water Resources, Div. Envir. Services, CA, USA
Cloern J, USGS, San Francisco CA, USA
Iriarte J, Instituto de Acuicultura- UACh, Chile
Many of the records in the data sets were only identified to
genera and were listed as sp., or spp., or by genus with a size class
designation such as “Gynmnodinium 20 mm” (see Olenina et al.,
2006) and consequently they were not used in this study. We did
not include data from Leblanc et al. (2012) since some of their
biovolume data did not come from a specific coastal site, but rather
as individual species records from taxonomic keys such as Tomas
(1997) and Kraberg et al. (2010), whereas the datasets of speciesspecific cell volumes that we used are from specific, globally
distributed coastal sites. However, we did compare the biovolume
of species that were common to all 3 datasets.
To the best of our knowledge, all biovolume measurements were
made according to the usual protocols. They were made on preserved cells, generally in Lugol's or formalin and the dimensions of
cells were measured microscopically assuming cells approximated
certain geometrical shapes following Hillebrand et al. (1999) or Sun
and Liu (2003). However, we were unable to determine if all of the
investigators used the same geometrical shape for the same species
due to the lack of metadata. Typically 10 to 50 cells were measured,
but the sampling time of year was not always available. We eliminated freshwater and benthic species, and cyanobacteria from our
analyses. The varieties, when indicated, were truncated and were
thus included in the parent species size calculation. Because the size
estimates were mostly based on fixed material, identification to the
species or even the genus level for naked phytoplankton <10 mm is
most likely poor. Therefore, our list is biased toward cells >10 mm.
2.2. Calculations and statistical analysis
We used the statistical analysis software package SAS 9.4™ to
standardized and merge data from the various sites and in the
analysis of statistical variables. In most of the 35 data sets, cell
volume was given so we estimated the equivalent spherical
diameter (ESD) from the volume (V) using Eq. (1), which is the
reverse of the more familiar equation for a sphere. However, in a
few data sets, only ESD was available and then volume was estimated from Eq. (2).
ESD ¼ 2
3 V
4 p
1
3
(1)
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 4/13
4
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
Table 2
Species volume (median, minemax, and mean with the % coefficient of variation (CV)) and median equivalent spherical diameter (ESD) aggregated over the 35 sites for 113
species of diatoms, 80 species of dinoflagellates and 21 others belonging to 13 classes that were included in 5 or more sites. The median volume was converted to carbon per cell
using the equations from Menden-Deuer and Lessard (2000). See methods for details. A tilde (~) in the first column follows species names which actually may include several
cryptic or pseudo-cryptic species.
Species Bacillariophyceae
No. of sites
ESD median
Actinoptychus octonarius
Actinoptychus senarius
Asterionellopsis glacialis
Asterolampra marylandica
Attheya septentrionalis
Aulacoseira distans
Bacteriastrum furcatum
Bacteriastrum hyalinum
Bellerochea horologicalis
Biddulphia rhombus
Cerataulina pelagica
Chaetoceros affinis
Chaetoceros borealis
Chaetoceros brevis
Chaetoceros castracanei
Chaetoceros circinalis
Chaetoceros compressus
Chaetoceros constrictus
Chaetoceros convolutus
Chaetoceros costatus
Chaetoceros curvisetus
Chaetoceros danicus
Chaetoceros debilis
Chaetoceros decipiens
Chaetoceros densus
Chaetoceros diadema
Chaetoceros didymus
Chaetoceros diversus
Chaetoceros eibenii
Chaetoceros laciniosus
Chaetoceros lauderi
Chaetoceros lorenzianus
Chaetoceros minimus
Chaetoceros muelleri
Chaetoceros neogracilis
Chaetoceros peruvianus
Chaetoceros radicans
Chaetoceros similis
Chaetoceros simplex
Chaetoceros socialis
Chaetoceros subtilis
Chaetoceros tenuissimus
Chaetoceros teres
Chaetoceros throndsenii
Chaetoceros tortissimus
Chaetoceros wighamii
Corethron criophilum
Corethron hystrix
Coscinodiscus centralis
Coscinodiscus concinnus
Coscinodiscus gigas
Coscinodiscus granii
Coscinodiscus oculus-iridis
Coscinodiscus radiatus
Coscinodiscus wailesii
Cyclotella choctawhatcheeana
Cyclotella striata
Cylindrotheca closterium
Dactyliosolen fragilissimus
Detonula confervacea
Detonula pumila
Ditylum brightwellii
Entomoneis paludosa
Eucampia zodiacus
Guinardia delicatula
Guinardia flaccida
Guinardia striata
Hemiaulus hauckii
Hemiaulus sinensis
Lauderia annulata
5
10
27
5
6
5
8
10
6
8
17
18
6
10
5
5
17
11
7
9
21
14
15
19
10
12
20
6
7
10
7
14
6
5
7
12
6
10
6
20
10
5
12
5
5
9
5
7
8
8
5
10
7
16
8
7
9
26
16
5
13
19
6
16
18
15
17
9
5
14
39.6
34.1
10
21.2
5.7
8.1
28.1
27.4
37.3
51.1
25.5
16.7
17.3
17.2
20.6
23.5
14
17.3
18.5
15
18.3
14.3
14.5
25.8
25.3
19.9
15.7
9.3
29.3
17.4
21.5
24
4.85
7.62
8.5
22.4
12.3
11.3
10.7
7.56
7.01
3.66
23.8
3.63
13.5
10.2
30.3
37.4
130
205
20.3
77
78.7
59.3
220
7.25
25
8.2
22.6
11.6
29.3
50.2
30.2
25.3
20.6
67.8
39.4
27.3
34.3
35.5
Volume median
32,400
20,700
524
4990
96.8
278
11,700
10,800
27,300
69,700
8720
2440
2720
2660
4590
6770
1450
2710
3320
1770
3220
1520
1600
9020
8480
4130
2030
421
13,200
2740
5180
7260
59.8
232
321
5860
984
764
643
226
180
25.7
7080
25
1290
562
14,600
27,500
1,160,000
4,540,000
4390
239,000
255,000
109,000
5,580,000
200
8180
289
6070
828
13,100
66,400
14,400
8460
4580
164,000
31,900
10,700
21,100
23,500
Volume minemax
6430e218,000
933e135,000
94e11,300
2100e442,000
31e211
148e905
1560e41,000
1780e105,000
1360e523,000
7790e440,000
2840e102,000
240e20,400
722e12,100
1780e15,800
2850e8950
2370e14,100
300e4630
1040e9700
2100e6190
507e6290
227e10,800
113e5080
403e48,500
155e57,700
1350e47,100
1420e27,400
97e20,400
218e540
1360e172,000
226e17,000
2130e11,900
262e43,800
21e3380
167e2810
48e1180
1830e58,900
221e3420
413e3700
155e2570
33e3150
45e15,900
25e33
2360e20,600
14e46
200e13,700
270e1770
6920e41,500
2470e1,420,000
257,000e3,330,000
143,000e61,600,000
442e61,300,000
93,600e2,240,000
3520e4,410,000
10,400e3,440,000
418,000e14,100,000
107e1140
552e43,800
10e5230
1680e143,000
498e2140
2360e43,700
28,100e274,000
4430e2,510,000
737e31,800
831e26,600
16,600e349,000
2770e201,000
7050e20,900
5100e48,500
13,500e137,000
Volume mean
(ESD CV%)
41,700(47)
19,600(42)
797(49)
45,800(81)
83(24)
389(26)
12,300(38)
17,100(43)
38,900(77)
105,000(49)
16,100(39)
3000(31)
3280(35)
4100(28)
4750(15)
7310(21)
1480(27)
2790(22)
3730(12)
2290(34)
2760(32)
1360(33)
2980(51)
9100(41)
8700(42)
4730(30)
2110(43)
374(13)
23,400(56)
3550(45)
5140(18)
7830(37)
194(81)
461(47)
296(39)
9070(42)
1080(37)
1070(28)
690(32)
356(45)
450(84)
27(4)
7690(23)
28(15)
1850(57)
663(24)
17,000(22)
101,000(78)
1,070,000(27)
5,480,000(60)
990,000(166)
422,000(42)
505,000(78)
130,000(62)
4,870,000(39)
320(33)
7580(50)
425(54)
8300(48)
922(19)
12,100(28)
73,400(22)
79,500(107)
9150(28)
5970(30)
134,000(26)
35,700(35)
10,800(12)
19,500(30)
27,100(24)
Cell carbon
(pg C cell1)
1310
911
46.2
287
11.7
27.7
573
536
1140
2440
452
161
176
173
269
368
105
175
206
124
201
110
114
465
442
247
139
38.7
633
177
296
389
7.95
23.9
31.1
328
77
62.8
54.5
23.4
19.5
4.01
382
3.92
95.9
48.9
685
1150
23,900
72,200
259
6620
6990
3510
85,400
21.1
429
28.5
337
66.9
630
2340
677
441
268
4870
1290
533
924
1010
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 5/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
5
Table 2 (continued )
Species Bacillariophyceae
No. of sites
ESD median
Leptocylindrus danicus (e)
Leptocylindrus mediterraneus
Leptocylindrus minimus
Lithodesmium undulatum
Melosira moniliformis
Melosira nummuloides
Melosira varians
Neocalyptrella robusta
Nitzschia acicularis
Nitzschia longissima
Odontella aurita
Paralia sulcata
Plagiogrammopsis vanheurckii
Porosira glacialis
Proboscia alata
Pseudo-nitzschia delicatissima (e)
Pseudo-nitzschia pseudodelicatissima
Pseudo-nitzschia pungens (e)
Pseudo-nitzschia seriata (e)
Pseudosolenia calcar-avis
Rhizosolenia hebetata
Rhizosolenia imbricata
Rhizosolenia setigera
Rhizosolenia styliformis
Skeletonema costatum (e)
Stephanopyxis turris
Thalassionema frauenfeldii
Thalassionema nitzschioides
Thalassiosira angulata
Thalassiosira anguste-lineata
Thalassiosira baltica
Thalassiosira eccentrica
Thalassiosira gravida
Thalassiosira minima
Thalassiosira nordenskioeldii
Thalassiosira pseudonana
Thalassiosira punctigera
Thalassiosira rotula
Thalassiothrix longissima
Trieres chinensis
Trieres mobiliensis
Trieres regia
Trigonium alternans
Akashiwo sanguinea
Alexandrium minutum
Alexandrium pseudogonyaulax
Alexandrium tamarense (e)
Amphidinium acutissimum
Amphidinium longum
Amylax triacantha
Archaeperidinium minutum
Dinophysis acuminata
Dinophysis acuta
Dinophysis caudata
Dinophysis fortii
Dinophysis norvegica
Diplopsalis lenticula
Dissodinium pseudolunula
Fragilidium subglobosum
Gonyaulax digitalis
Gonyaulax monacantha
Gonyaulax polygramma
Gonyaulax spinifera
Gonyaulax verior
Gymnodinium simplex
Gymnodinium verruculosum
Gyrodinium fusiforme
Gyrodinium spirale
Heterocapsa rotundata
Heterocapsa triquetra
Karlodinium veneficum
Katodinium glaucum
Lepidodinium chlorophorum
25
10
21
8
8
8
9
11
8
14
13
16
5
6
17
15
6
11
10
8
9
14
22
14
25
6
12
28
8
11
6
12
6
5
16
7
7
17
6
14
11
6
7
15
6
5
9
7
5
9
8
15
8
8
7
5
7
5
5
11
5
6
10
7
8
5
9
7
14
15
6
9
10
15
25.1
7.07
40.2
35.2
22.7
20.6
147
8.42
9.75
32.6
21.3
8.92
53.2
32.6
7.25
7.26
13
13.5
73.3
52
43.2
27.4
59.6
7.44
31.6
14.1
11.4
27.4
34.2
43.2
32.6
29.5
8.23
24.1
4.7
55
28.6
24.8
118
47.3
110
27.1
40
22.3
42
29.7
9.76
14
27.8
30.6
27.3
47
44
43.6
39.4
39.6
35.4
42.3
32.4
36.7
31.3
31
24.9
9.04
11.1
30.3
34.1
7
16.6
11.4
15.6
18.3
Volume median
1750
8250
185
34,000
22,900
6110
4590
1,660,000
312
485
18,200
5030
372
78,900
18,200
200
200
1140
1290
206,000
73,400
42,200
10,800
111,000
215
16,500
1460
774
10,700
21,000
42,100
18,100
13,500
291
7330
54.3
87,100
12,200
8000
864,000
55,300
692,000
10,400
33,500
5790
38,800
13,700
487
1440
11,300
15,000
10,700
54,300
44,500
43,300
32,000
32,500
23,200
39,500
17,800
25,900
16,100
15,600
8080
387
716
14,500
20,700
180
2400
772
1980
3190
Volume minemax
95e10,400
382e68,200
10e3380
9780e108,000
4190e39,100
905e22,300
1360e114,000
75,400e15,800,000
14e2680
24e15,400
919e148,000
1150e68,000
257e318,000
19,000e180,000
281e306,000
53e507
48e271
66e4320
300e3550
38,400e2,750,000
439e1,88,000
1360e1,560,000
1480e1,50,000
18,900e1,290,000
50e1810
6840e47,700
299e6970
50e143,000
3170e24,000
7750e65,400
22,100e80,200
162e202,000
2920e27,600
185e524
193e499,000
28e200
12,800e175,000
3450e47,000
2150e98,200
9280e10,400,000
4900e439,000
8080e5,510,000
580e90,500
2490e231,000
2260e15,700
36,800e47,400
3370e26,400
206e1050
697e2850
4600e16,800
25e41,600
1670e40,700
1580e105,000
19,500e196,000
368e56,800
25,500e121,000
10,800e106,000
789e38,000
14,100e52,300
2950e43,400
12,200e109,000
10,600e41,700
10,700e23,800
7120e39,300
203e1700
322e3650
3520e109,000
3160e113,000
115e290
965e5700
140e1130
416e18,300
1640e15,000
Volume mean
(ESD CV%)
2330(30)
10,200(51)
312(53)
41,100(26)
18,600(26)
5600(35)
10,300(54)
2,800,000(58)
372(48)
599(63)
20,200(48)
6930(43)
8460(132)
76,600(25)
21,500(55)
199(22)
176(18)
1130(30)
1370(24)
361,000(60)
57,100(46)
83,000(67)
18,000(44)
176,000(48)
378(35)
18,100(22)
1500(35)
1240(80)
9850(25)
22,300(20)
44,400(18)
18,500(68)
13,000(26)
307(15)
15,400(79)
62(27)
73,700(31)
12,000(25)
13,500(53)
991,000(58)
72,000(42)
653,000(65)
14,300(45)
32,600(40)
6060(24)
41,500(4)
12,700(23)
502(20)
1580(17)
10,900(13)
14,500(40)
10,600(26)
43,800(29)
53,300(24)
30,500(35)
42,000(24)
34,400(24)
15,800(41)
34,400(18)
17,600(21)
30,600(31)
18,800(17)
16,100(9)
11,900(23)
463(26)
887(36)
18,100(35)
25,700(38)
177(12)
2460(16)
645(22)
2660(38)
4450(28)
Cell carbon
(pg C cell1)
123
432
19.9
1360
990
339
269
32,000
30.4
43.4
821
289
35
2700
821
21.1
21.2
86.9
95.9
5880
2540
1620
536
3560
22.5
759
106
63.4
535
922
1620
819
643
28.7
393
7.35
2920
595
422
18,800
2020
15,700
522
3830
737
4400
1660
72.1
199
1380
1810
1310
6030
5000
4870
3670
3730
2720
4480
2120
3010
1920
1870
1010
58.1
104
1750
2440
28.3
322
111
269
421
(continued on next page)
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 6/13
6
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
Table 2 (continued )
Species Bacillariophyceae
No. of sites
ESD median
Volume median
Volume minemax
Volume mean
(ESD CV%)
Cell carbon
(pg C cell1)
Lessardia elongata
Lingulodinium polyedrum
Mesoporos perforatus
Micracanthodinium setiferum
Nematopsides vigilans
Noctiluca scintillans
Oblea rotunda
Oxyrrhis marina
Oxytoxum variabile
Peridiniella catenata
Peridiniella danica
Phalacroma rotundatum
Polykrikos schwartzii
Pronoctiluca pelagica
Prorocentrum balticum
Prorocentrum cordatum
Prorocentrum gracile
Prorocentrum micans
Prorocentrum triestinum
Protoceratium reticulatum
Protoperidinium bipes
Protoperidinium breve
Protoperidinium brevipes
Protoperidinium claudicans
Protoperidinium conicoides
Protoperidinium conicum
Protoperidinium crassipes
Protoperidinium denticulatum
Protoperidinium depressum
Protoperidinium divergens
Protoperidinium granii
Protoperidinium leonis
Protoperidinium oblongum
Protoperidinium pallidum
Protoperidinium pellucidum
Protoperidinium pentagonum
Protoperidinium pyriforme
Protoperidinium steinii
Pyrophacus horologium
Scrippsiella trochoidea (e)
Torodinium robustum
Tripos furca
Tripos fusus
Tripos lineatus
Tripos longipes
Tripos macroceros
Tripos muelleri
Tripos pentagonum
Tripos trichoceros
Tripos uncinus
Hemiselmis virescens (Cryptophyceae)
Komma caudata (Cryptophyceae)
Plagioselmis prolonga (Cryptophyceae)
Rhodomonas marina (Cryptophyceae)
Teleaulax acuta (Cryptophyceae)
Teleaulax amphioxeia (Cryptophyceae)
Apedinella radians (Dictyochophyceae)
Dictyocha fibula (Dictyochophyceae)
Pseudopedinella pyriformis (Dictyochophyceae)
Ebria tripartita (Ebriophyceae)
Eutreptia lanowii (Euglenoidea)
Leucocryptos marina (Katablepharidophyta)
Micromonas pusilla (Mamiellophyceae)
Pseudoscourfieldia marina (Prasinophyceae)
Pyramimonas amylifera (Prasinophyceae)
Paulinella ovalis (Proteomyxidea)
Acanthoica quattrospina (Prymnesiophyceae)
Emiliania huxleyi (Prymnesiophyceae)
Pleurochrysis carterae (Prymnesiophyceae)
Syracosphaera pulchra (Prymnesiophyceae)
Heterosigma akashiwo (Raphidophyceae)
6
8
6
5
6
8
8
7
6
6
9
12
8
6
9
15
9
21
12
7
11
7
10
7
6
14
5
5
16
10
7
7
6
10
11
8
5
9
8
14
10
22
20
16
5
7
16
5
5
10
7
8
6
7
6
10
11
12
10
11
6
9
5
5
5
8
5
13
5
6
8
10.8
40.9
15.5
15.8
21
586
25.4
15.9
8.14
24.7
17.3
30.4
66.6
18.3
12
12.1
19.1
26.5
12.9
30.9
16.1
33.1
27.9
60.3
45.6
57.4
70
30
95.4
60.7
44
62.1
56.5
54.2
37.6
53.3
40
31.2
43.8
20.7
18
46.9
33.4
29.7
48
44.5
51.3
41.5
37
50.1
3
5.3
4.53
7.46
7.36
6.91
7.27
21.4
6.6
26.9
11.6
6.13
1.41
2.44
9.81
3.15
7.5
6
14.5
15.8
15.4
657
35,800
1940
2050
4840
105,000,000
8610
2090
282
7850
2710
14,700
155,000
3200
914
921
3650
9700
1130
15,400
2200
19,100
11,300
115,000
49,700
99,000
180,000
14,100
454,000
117,000
44,600
125,000
94,500
83,400
27,900
79,100
33,500
16,000
44,000
4630
3070
54,200
19,400
13,700
58,000
46,100
70,900
37,400
26,600
65,900
14.1
77.9
48.6
217
209
173
201
5130
151
10,200
817
121
1.46
7.61
494
16.4
221
113
1600
2050
1910
359e1070
18,600e65,800
1150e3940
1020e3480
3890e7390
860,000e382,000,000
4190e14,100
268e6370
78e476
5700e92,400
595e5810
499e116,000
27,300e293,000
905e10,500
20e1590
237e2350
49e19,700
71e18,600
335e3660
7450e31,300
432e3650
6880e36,900
5450e14,900
9400e156,000
10,700e103,000
6060e345,000
81,300e606,000
14,100e49,200
21,900e1,330,000
67,400e1,410,000
15,800e86,600
41,800e260,000
49,700e231,000
2620e244,000
6800e78,800
4290e572,000
12,200e108,000
9160e58,900
8890e182,000
690e16,800
1770e10,600
3000e123,000
6260e283,000
1110e96,300
43,300e139,000
3730e69,100
414e814,000
30,200e57,500
1160e106,000
1950e139,000
12e3690
48e283
42e64
113e1240
80e278
65e386
141e386
345e55,800
100e180
2520e14,100
284e2680
84e910
1e5
3e14
145e589
10e318
196e335
34e213
630e3060
871e8680
89e4190
637(15)
36,500(13)
2110(17)
2070(16)
5030(7)
83,700,000(51)
8060(14)
2070(34)
249(20)
13,900(44)
2430(23)
17,400(41)
139,000(28)
3760(29)
633(37)
926(19)
4900(44)
7170(31)
1250(23)
15,900(15)
2170(18)
18,100(20)
10,100(12)
85,200(26)
44,100(25)
99,400(28)
207,000(29)
21,700(20)
377,000(29)
156,000(39)
37,600(19)
117,000(21)
105,000(22)
71,000(33)
29,700(22)
99,500(44)
38,400(27)
19,600(21)
43,700(39)
4730(28)
3510(19)
47,200(26)
29,500(40)
13,800(32)
69,400(17)
34,700(30)
63,200(54)
41,100(9)
18,600(60)
54,200(27)
114(101)
104(25)
50(6)
381(31)
187(14)
177(22)
226(11)
7050(44)
141(8)
8360(17)
915(25)
218(30)
2(28)
8(21)
411(17)
31(49)
242(8)
100(20)
1400(22)
2850(31)
1580(34)
95.6
4080
264
278
623
316,000
1070
283
43.2
981
362
1770
16,100
422
130
131
478
1200
159
1850
297
2260
1390
12,200
5550
10,600
18,500
1700
44,300
12,400
5020
13,200
10,200
9030
3230
8590
3830
1910
4950
597
407
6020
2300
1650
6420
5170
7750
4250
3090
7230
2.6
12.9
8.28
33.8
32.6
27.2
31.4
658
24
1260
117
19.4
0.308
1.45
73.1
2.99
34.3
18.3
220
279
260
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 7/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
7
Table 3
Genus-specific volume (median, minemax and mean and % coefficient of variation (CV)) and median ESD, aggregated over the species listed in Table 2, for the 10 genera of
diatoms and 6 genera of dinoflagellates having more than 3 species listed in Table 2.
Genus
Bacillariophyceae
Chaetoceros
Coscinodiscus
Guinardia
Leptocylindrus
Melosira
Pseudo-nitzschia
Rhizosolenia
Thalassiosira
Trieres
Dinophyceae
Alexandrium
Dinophysis
Gonyaulax
Prorocentrum
Protoperidinium
Tripos
No. of species
ESD median
Volume median
Volume minemax
Volume mean (ESD CV%)
35
7
3
3
3
3
4
10
3
15.7
78.7
39.4
15
22.7
13
47.6
29.05
110
2030
25,5000
32,000
1770
6120
1150
56,500
12,800
697,000
25e13,200
4380e5,580,000
4580e163,000
185e8280
4580e22,800
200e1290
10,800e111,000
54e87,100
55,400e860,000
2760(6.7)
990,000(75)
35,700(24)
2330(9.0)
10,300(7.9)
1130(3.5)
69,300(14)
14,100(15)
653,000(39)
3
5
5
5
17
9
29.7
43.6
31.3
12.9
53.3
44.5
13,700
43,400
16,100
1120
79,300
46,100
4
ESD 3
V¼ p
3
2
(2)
Some of the species size estimates for a site were obtained in the
form of a single value from either the literature or from the data
contributor (Table 1). However, in other cases (e.g. North Western
Atlantic and the Baltic Sea), cell sizes are estimated seasonally and
in some cases from multiple sampling stations collected over many
years (Denmark, Finland and Sweden). In these cases, we first
estimated a monthly average of equivalent spherical diameter
(ESD) that was in turn averaged to give the ESD for each species for
a site.
For statistical purposes, we considered only the 214 species that
appeared at least 5 times in the 35 data sets. The distribution of
volumes was skewed right so all statistical calculations were performed on the more normally-distributed equivalent spherical
diameter (ESD), as recommended by Hillebrand et al. (1999).
Working with diameters, this method is equivalent to using a cuberoot transformation of volume, which is similar in effect to the
more familiar log transformation applied to log-normally distributed data. For these 214 species in the 35 data sets, we calculated
the median ESD and its corresponding volume, the range (minemax) in ESD expressed as volume, the mean ESD expressed as
volume, and the coefficient of variation (CV) of the ESD across the 5
or more sites. Table 2 lists the resulting values of ESD and cell
volume for each species, and also the carbon per cell obtained by
multiplying the median volume by the carbon density relation as
recommended by Menden-Deuer and Lessard (2000). The median
volume was converted to carbon per cell using the equation C ¼ aVb
where a and b are 0.288 and 0.811 for diatoms, 0.216 and 0.939 for
other protists, and 0.003 and 1 for Noctiluca scintillans. After
aggregating over sites in Table 2, we then aggregated over species
5810e38,800
10,700e54,400
8080e25,900
905e9740
11,400e455,000
13,700e70,700
12,700(10)
42,000(7.7)
17,600(4.2)
1250(6.3)
71,000(17)
41,100(7.6)
to obtain ESD and volume statistics for the most common genera
for diatoms and dinoflagellates with 3 or more species (Table 3). We
also aggregated over species for the 5 dominant classes that had 3
or more genera (Table 4). Finally, we compared our results from
field samples to those from laboratory cultures and culture collections (Table 5).
The minimum and maximum ESD obtained in this study was
compared to the minimum and maximum ESD obtained by Olenina
et al. (2006) and Leblanc et al. (2012) using a type II linear
regression (uncertainty associated with both axes). Because most of
the observed cells were small, we log-transformed ESD for both
axes to distribute the observations equally along the length scale
and we assumed that the uncertainty of the estimate of ESD scaled
with the increase in ESD of the cell. Therefore, the regression was
carried out by formulating a linear model for ESD (log Olenina
min ¼ a þ b log This Study min, etc.), but fitting the model to
log(ESD) using a non-linear regression. We also tested whether the
slope of the different type II regressions of minimum and maximum
values of Olenina et al. (2006) and Leblanc et al. (2012) scaled with
a slope b that was significantly different from 1 (b ¼ 1) using the
Wald test of SAS 9.4. Briefly, Wald's statistic is a test statistic with a
known probability distribution (a chi-square distribution) and can
be used to test whether the slope in a type II regression model is
significantly different from zero.
3. Results
Table 1 lists the general location of the 35 global coastal sites
that were grouped into regions. Four sites were from the western
North Atlantic, and 15 were from the eastern N Atlantic region,
predominately from the Baltic, North Sea and the Mediterranean
Sea. Only 2 sites were from the S Atlantic off Chile. Five sites were
from coastal sites in the Indian Ocean. The east coast of the Pacific
Table 4
Class-specific volume (median, minemax and mean and % coefficient of variation (CV)) and median ESD, aggregated over the species listed in Table 2, for classes having more
than 3 genera.
Class
No. of species
ESD median
Volume median
Volume minemax
Volume mean (ESD CV%)
Bacillariophyceae
Cryptophyceae
Dictyochophyceae
Dinophyceae
Dinophyceae (excluding Noctiluca)
Prymnesiophyceae
114
6
3
80
79
4
25.4
6.49
7.56
32.4
32.3
10.8
8580
143
226
17,900
17,600
660
27e5,480,000
50e381
141e7050
177e83,700,000
177e377,000
100e2850
22,600(107)
149(23)
1050(77)
32,400(150)
19,100(50)
743(49)
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 8/13
8
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
Table 5
Comparison of biovolumes of the same species that appeared in Table 2 (column 1) with the same species grown in laboratory cultures by various investigators. The biovolumes
that are in parentheses in column 4 and 5 are for Lugol's fixed compared to live volumes. Column 1 ¼ from Table 2; 2 ¼ Kim et al. (2013); 3 ¼ Menden-Deuer and Lessard
(2000); 4 ¼ Menden-Deuer et al. (2001); 5 ¼ Montagnes et al. (1994); 6 ¼ Montagnes and Franklin (2001); 7 ¼ Chan (1978).
Diatoms
Chaetoceros didymus
C. simplex
Detonula pumila
Ditylum brightwellii
Leptocylindrus danicus
Skeletonema costatum sl
Thalassiosira eccentrica
T. pseudonana
T. gravida
Dinoflagellates
Akashiwo sanguinea
Gymnodinium simplex
Heterocapsa triquetra
Lingulodinium polyedrum
Prorocentrum micans
Protoperidinium conicum
P. depressum
P. pellucidum
Scrippsiella trochoidea
Tripos furca
Tripos fusus
Tripos lineatus
1
2
2110
690
12,100
73,400
2330
378
18,500
62
13,000
32,600
463
2460
36,500
7170
99,400
377,000
29,700
4730
47,200
29,500
13,800
3
4
5
6
7
1062
160
7395(6390)
8230
9713 (7291)
37 (25)
16,400
42
200
60,000
31
637
42 (30)
14,273 (13,170)
88,099
180
660
28,300
17,300
4080
350
25,000
16,303
50,721
278,883
105,667
8474
47,435
was represented by 4 sites along Canada and the USA and the
eastern Pacific, and the western Pacific had one site from Korea, and
3 from China. There were 32 sites in the Northern Hemisphere and
only 3 sites were from the Southern Hemisphere. Most of the sites
were temperate with only 2 tropical sites.
Among the 8760 records, there were 214 unique marine
phytoplankton species that occurred five or more times and consisted of 113 diatoms, 80 dinoflagellates and 21 others belonging to
13 classes (Table 2). We included Protoperidinium and other heterotrophic flagellates commonly recorded in a phytoplankton
sample. The genus Chaetoceros spp. dominated the diatoms with 35
species, followed by 10 species of Thalassiosira. In the 80 dinoflagellates that were listed, there were 17 species of Protoperidinium, and 9 species of Tripos (previously called Ceratium).
The median ESD and median volumes of different diatom genera
varied by ~60 times and ~200,000 times respectively. The smallest
ESDs of 3.7 and 3.6 mm and the corresponding smallest biovolumes
of 26 and 25 mm3 were reported for Chaetoceros tenuissimus and
Chaetoceros throndsenii respectively (Table 2). In contrast, the
largest ESDs of 220 and 205 mm and the corresponding biovolumes
of 5.6 and 4.5 106 m m3 were for Coscinodiscus wailesii and
Coscinodiscus concinnus respectively.
The largest diatoms had about 20,000 times more carbon per
cell than the smallest ones (Table 2). For the two smallest Chaetoceros sp., carbon per cell was ~4 pg C cell1. In contrast, for the two
large Coscinodiscus species, carbon per cell was ~8 104 pg C cell1.
For the two smallest Chaetoceros species, the carbon density (or
carbon per unit cell volume) was ~0.16 pg C mm3, while for the two
large
Coscinodiscus
species
the
carbon
density
was
~0.016 pg C mm3. Therefore, the carbon density of these two small
diatoms was ~10 times greater than the two largest diatoms.
The median ESD and median volumes for dinoflagellate genera
varied by ~10 and ~1500 times, respectively (Table 2). The lowest
ESDs of 7 and 8.1 mm and the smallest biovolumes of 180 and
282 mm3 were for Heterocapsa rotundata and Oxytoxum variabile,
respectively (Table 2). In contrast, the largest ESDs of 95 and 67 mm
and biovolumes of 4.5 and 1.6 105 mm3 were for Protoperidinium
depressum and Polykrikos schwartzii, respectively.
The largest dinoflagellates had about 1000 times more carbon
per cell than the smallest ones. The lowest carbon per cell
71,859 (55,045)
209 (218)
1795 (1914)
34,663 (30,167)
379 (224)
400
10,000
454,451 (578,364)
4408 (4873)
46,514
44,619 (54,687)
9000
(28e43 pg C cell1) was for the two small dinoflagellates listed
above. The highest carbon per cell (1.6 and 4.4 104 pg C cell1)
was for the two large species above (Table 2). For the two small
dinoflagellates, the carbon density was ~0.15 pg C mm3, while for
the two large dinoflagellates, the carbon density was lower at
0.10 pg C mm3. Therefore, large cells of dinoflagellates have 50%
lower carbon density that small dinoflagellates. Small cells of diatoms and dinoflagellates had 10 and 1.5 times respectively greater
carbon density than large cells and clearly indicting the large
variation in diatoms compared to dinoflagellates.
For the 21 species in the ‘others’ category that were composed of
9 classes, ESDs were relatively small, ranging from 1.4 to 27 mm and
the biovolumes ranged from a low of 1.5e~10,200 mm3 for Micromonas pusilla and Ebria tripartita, respectively (Table 2). There was a
large range in carbon per cell from 0.2 to 1260 pg C cell1, but a very
small difference in the carbon density (0.137e0.123 pg C mm3) for
the small and large species respectively, which is markedly lower
than the carbon density for small cells of diatoms (0.16 pg C mm3)
and dinoflagellates (0.15 pg C mm3).
The chrysophytes had the smallest ESD of 6.5 mm and the
smallest median volume of 143 mm3 of the 5 classes (Table 4). The
dinoflagellates had the largest ESD of 32 mm and median volume of
17,600 mm3. The mean ESD for all 113 diatoms was 25.5 mm, with
median volume of 8700 mm3 and CV of 106% (Table 4).
Larger species diatoms, tend to have significantly lower carbon
density than dinoflagellates (0.016 vs. 0.1 pg C mm3). The mean
ESD for all 113 diatoms was 25.5 mm, with median volume of
8700 mm3 and CV of 106% (Table 4). Data for many species tend to fit
the curve C/V ¼ a V(b1), with the coefficient a being equal to the
carbon density at V ¼ 1, and the coefficient b describing how carbon
density varies with volume. For example, for diatoms, MendenDeuer and Lessard (2000) recommended a ¼ 0.288 and b ¼ 0.811.
With a 10-fold increase in cell volume, the carbon density reduces
to 65% of its original value (10(10.811) ¼ 0.65). For other protists,
a ¼ 0.216 and b ¼ 0.939, so a 10-fold increase in size corresponds to
only a 15% reduction in carbon density. Clearly the change in carbon
density is minor compared to the range in biovolume for a species.
Considering the chain of propagation of errors, biovolume warrants
more attention than refining carbon density. This analysis is further
supported by Jakobsen et al. (2015), who also identified biovolumes
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 9/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
of diatoms as an important source of uncertainty in the estimation
of natural community carbon biomass. Furthermore, numeration
and identification errors may be even more important (Zingone
et al., 2015). It is worth noting that the well-documented relation
between carbon density and volume refers to many different species of widely varying sizes. But there is no evidence that the
relation applies to a single species that changes its size in response
to environmental conditions or life cycle stage.
Many species that are grown in the laboratory are obtained from
culture collections and they are often selected for their ease of
culturing and may not be ecologically important. However, there
were 9 diatoms and 15 dinoflagellates in Table 2 that have also been
used in previous laboratory experiments where their biovolume
was measured (Table 5). Some biovolumes were measured on live
samples and in two studies a live vs. fixed comparison was made
(Montagnes et al., 1994; Menden-Deuer et al., 2001). Comparing 9
diatoms from lab vs. field samples, there was relatively good
agreement (e.g. a factor of 2), but there was up to an order of
magnitude lower volumes for lab cultures of Ditylum brightwellii,
Leptocylindrus danicus and Thalassiosira gravida, due to strain differences in the same species (Menden-Deuer et al., 2001; Kim and
Menden-Deuer, 2013). There was less variation among laboratory
grown cultures for the dinoflagellates, except for Prorocentrum
micans where there was a 10-fold difference in the cell volume
between two different strains of P. micans that were used by
Menden-Deuer & Lessard (2000) and Menden-Deuer et al. (2001).
These examples point out the importance of including information
on the strain that is used in lab experiments.
The range (minemax) of ESD for 84 diatom species that were
found in all three data sets (this study, Olenina et al. (2006) and
Leblanc et al. (2012) indicates good general agreement between
these 3 data sets despite the pronounced differences in the
geographical distribution of these 3 data sources (Fig. 1). Similarly,
there is good general agreement for the 79 dinoflagellates in this
studywith those of Olenina et al. (2006) (Fig. 1). The type II linear
regression on all records of diatoms and dinoflagellates also confirms that there was generally good correspondence between the
minimum and maximum ESD estimated in our study and the ESD
estimated by Olenina et al. (2006) and Leblanc et al. (2012)
(Table 6). Only the minimum ESD scaled significantly lower by
2.7% than ESD obtained in this study (P ¼ 0.286), whereas ESD for
all the other comparisons, scaled with a slope of 1 (Table 6,
P < 0.001). The uncertainty associated with the individual species
minimum and maximum ESD estimates showed a reasonably low
uncertainty (48e62%) in the size estimates, except for the minimum ESD of Leblanc et al. (2012) and this study that had an uncertainty of ~82e83%. Thus, the globally distributed data in this
study matched the more local Baltic Sea data of Olenina et al.
(2006), whereas some discrepancy was found in the comparison
to the data of Leblanc et al. (2012).
4. Discussion
With the emphasis on long-term monitoring to assess changes
in water quality in coastal areas, there is renewed interest in
changes in phytoplankton species composition, abundance and
biomass as indicators of environmental impacts (e.g. European
Commission, 2000, 2008). At present, there are very few datasets
on biovolumes for a large number of species from field samples
from various coastal oceans. One of the best organized groups is the
HELCOM (Helsinki Commission) Phytoplankton Expert Group
(PEG) with members from the nine countries that border on the
Baltic Sea. They have set up standard protocols for collecting, preserving, counting and estimating biovolumes using agreed upon
geometrical shapes and size classes for various species (Olenina
9
66
et al., 2006) (HELCOM, 2015). With these standardized protocols
67
that are regularly being updated, the monitoring datasets from
68
various labs that are members of HELCOM PEG are now inter69
comparable and HELCOM countries have agreed to upload their
70
data to a common open database maintained by ICES. Recently,
71
three countries with phytoplankton monitoring stations in the
72
Black Sea have drawn up standardized protocols for biovolume
73
measurements that are similar to those used in the Baltic Sea
74
(Moncheva et al., 2014). Similarly, the European Committee for
75
Standardization (2015) conducted a European inter-laboratory
76
comparison and is developing a CEN standard protocol for the
determination of phytoplankton biovolumes. While developing Q3 77
78
regional protocols for biovolume measurements is a good begin79
ning, eventually a global standard protocol must be developed
80
(Jakobsen et al., 2015) with detailed quality controls (Rott et al.,
81
2007; Zingone et al., 2015), similar to QA/QC methods that are
82
used for chemical analysis of seawater (e.g. Parsons et al., 1984).
83
During the last decade, two significant studies have recom84
mended geometrical shapes for species and produced detailed
85
guidelines for biovolume calculations, taking into account the
86
complexity of numerous cell shapes (Hillebrand et al., 1999; Sun
87
and Liu, 2003). Hillebrand et al. (1999) presented 19 geometrical
88
shapes and geometrical equations for calculating biovolumes for
89
more than 850 pelagic and benthic marine and fresh water
90
microalgal genera, and Sun and Liu (2003) list 31 geometrical
91
shapes for 284 marine genera and 863 genera in their ‘Biovolume
92
Tools’ program. However, the European Committee for
93
Standardization (2015) is recommending reducing the number of
94
geometrical shapes, especially for those species with a complex
95
geometrical shape that requires several measurements of various
96
dimensions, in order to simplify the time-consuming measurement
97
process.
98
In some time series data sets, often identification is only to the
99
genus level. In this case, the species-specific volumes in Table 2
100
cannot be directly applied. The biovolumes for common genera in
101
Table 3 clearly indicate the very wide range in volumes and hence,
102
carbon biomass. It is recommended that for genera that are
103
important biomass contributors, taxonomic assistance be obtained
104
to identify cells to the species level which will improve the reli105
ability of the carbon biomass estimate, or at least note/photograph
106
their shape and dimensions.
107
Another important choice for inter-calibration with other
108
groups is to decide whether to measure cell volume on live or
109
preserved cells. Shrinkage varies with the type of preservative and
110
the fixative concentration. Montagnes et al. (1994) found that cells
111
shrink immediately when Lugol's preservative is used. Similarly,
112
Verity et al. (1992) found that gluteraldehyde caused cell volume to
113
shrink ~30% on average. In contrast, Menden-Deuer et al. (2001)
114
found that both swelling (up to 30%) and shrinkage (up to 60%)
115
was observed for dinoflagellates and some diatoms that were
116
preserved with Lugol's or gluteraldehyde and hence, they
117
concluded that the effect of preservation is species-specific.
118
Nevertheless, they concluded that fixation-induced volume
119
changes appear to be a negligible factor in estimates of field sam120
ples containing many species, but not in the case where one or two
121
species dominate the assemblage. Therefore, it is recommended
122
that the effect of the preservative that is used by the monitoring
123
program be determined for possible species-specific shrinkage of
124
the biovolume of at least the ecologically abundant phytoplankton
125
species in the study area.
126
Comparisons with the very large dataset from the HELCOM PEG
127
for the Baltic Sea (Olenina et al., 2006) with a large salinity gradient
128
from ~2 in the north to ~30 in the south, revealed that the range of
129
ESDs for diatoms and dinoflagellates in Table 2 usually fell within
130
the range of the ESDs for the Baltic Sea (Fig. 1). This suggests that
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 10/13
10
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
the influence of spatial variation among areas may be quite small
because the Baltic Sea dataset represents only a relatively small
spatial scale compared to datasets in Table 2 that are from different
oceans and hemispheres.
The HELCOM PEG uses up to 23 size classes to describe the size
variation in the diatom Skeletonema for example. Even though this
taxon includes several species, some of which are pseudo-cryptic or
cryptic (Kooistra et al., 2008), in this case, it has been confirmed by
Fig. 1. Left panel: Range (minemax) of equivalent spherical diameter (ESD) for 84 species of diatoms that were common to this study, Olenina et al. (2006), and Leblanc et al. (2012).
Species are sorted from top to bottom in order of the mean ESD. Right panel: Same but for dinoflagellates and only two studies (this study, with species matches from Olenina et al.
(2006)).
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 11/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
11
Table 6
Parameter estimates for the type II linear regression (y ¼ a þ bx) for fitting cell size (ESD) of Olenina et al. (2006) to the ESD for diatoms in this study, and testing the hypotheses
of zero intercept and a slope equal to one. The root mean squared error (RMSE) ESD for this study and RMSE of the minimal and maximal ESD of Olenina et al. (2006) and
Leblanc et al. (2012) give the RMSE for the log-transform of the two regression variables. Numbers in parentheses are the corresponding percent uncertainty (relative uncertainty (%) was estimated as (eRMSE 1) 100)).
Minimum ESD(Olenina et al. 2006)
Maximum ESD(Olenina et al. 2006)
Minimum ESD(Leblanc et al. 2012)
Maximum ESD(Leblanc et al. 2012)
a
P(a ¼ 0)
b
P(b ¼ 1)
RMSE
0.98
1.58
0.076
0.572
<0.0001
<0.0001
<0.839
<0.0001
0.917
1.07
0.973
0.94
<0.001
<0.001
<0.286
<0.001
0.484(62%)
0.400(49%)
0.599(82%)
0.457(58%)
extensive spatial sampling and molecular analysis that the only/
main Skeletomema species present in the Baltic Sea is S. marinoi
(Almany et al., 2009). Therefore, the large variation in the cell size of
Skeletonema in the Baltic Sea is not likely due to several cryptic
species, but more likely due to different environmental factors and
life cycles (Godhe et al., 2014). Olenina et al. (2006) used 16 size
classes for Cerataulina pelagica, 12 for Chaetoceros subtilis and 6 for
Heterocapsa triquetra, for example. The use of size classes has the
advantage that monthly or seasonal variations in cell size are
recorded during the counting process and this is perhaps more
practical than measuring species cell volumes on a monthly basis.
In order to facilitate inter-lab comparisons, it would be helpful to
have a recommended list of typical size classes for various species
which have significant variability in their size, similar to the recommendations for the geometric shapes for various species. There
were 52 species in Table 2 that were not in the Baltic Sea species list,
indicating that our global dataset has a more diverse assemblage of
species and therefore should be more broadly applicable for studies
in different coastal oceans.
At first glance, obtaining species-specific cell carbon estimates
from biovolume appears to be relatively simple. There are at least 7
steps in this process to consider. 1) Correct species identification.
This is determined by the expertise of the taxonomist which can
vary considerably; small cells are particularly challenging with a
light microscope (Jakobsen et al., 2015; Zingone et al., 2015). 2) Use
of preserved or live cells. Shrinkage is well known for many preserved species, and hence, the use of live cells for the most
ecologically important species in a study area is recommended, if
possible. 3) Cell dimensions. Measuring cell length and width are
straight forward (except for the halo effect), but measuring the
third dimension, the cell depth or height (the hidden dimension) is
more difficult (see Sun and Liu, 2003 for suggestions). 4) Number of
cells measured. Previous tests have shown that 10 to 50 cells should
be measured, depending on the size and complexity of the shape of
the cell. 5) Use of size classes. Size varies >10-fold for some species,
especially diatoms, and therefore the use of size categories helps to
capture the well-known seasonal variation in biovolume. If size
classes are not used, then monthly biovolume should be determined for the most important species in the study site. Diatom cell
volumes are the single largest source of uncertainty in community
carbon estimates and exceeds the uncertainty associated with the
different volume to carbon estimates (e.g. Montagnes et al., 1994;
Menden-Deuer and Lessard, 2000; Jakobsen et al., 2015). Monthly
biovolumes help to document the influence of simultaneous environmental and biological factors such as light, temperature, nutrients, salinity and life cycles. 7) Cell carbon to biovolume conversion
factors. Only about 30e50 species (some of which are laboratory
cultures that are easy to grow and not necessarily ecologically
important), have been used to produce the regressions that are
presently used for carbon and biovolume (Montagnes et al., 1994;
Menden-Deuer and Lessard, 2000). Hence, there is a need for
more ecologically relevant species with a large size range to be
examined by growing them in the lab and directly measuring their
this study
RMSE Olenina et al. (2006)/Leblanc et al. (2012)
0.435
0.396
0.605
0.449
(55%)
(48%)
(83%)
(57%)
cell carbon and size. Considering the uncertainty and possible errors in each of these steps above, the species-specific carbon
biomass estimates from this study should be used cautiously since
they may easily vary by a factor of 2 to >10 times.
The range in biovolume estimates in this paper includes the
variation due to all of the steps above. There must be considerable
variation in the expertise of the many different taxonomists who
determined the biovolumes along with the uncertainty of the
reliability of the species identifications. Mostly preserved cells were
measured in the data sets. Probably the recommended geometric
shapes were not used in all cases and there was no information on
what time of the year that the volumes were measured. Nevertheless, this global dataset of biovolumes is useful in order to
evaluate where locally determined biovolumes lie within the global
spectrum of spatial and temporal variations. In addition, the dataset
could be widely used to convert cell numbers into biomass when
local biovolume measurements are missing, which would confer
some homogeneity to the carbon biomass data obtained. Finally,
this study highlights the need to adopt standard protocols for
measuring biovolumes and preferably on a monthly basis especially
for the dominant diatoms in the study area. However, since
monthly volume measurements are very time consuming, binning
the cells into size classes can capture changes in size/volume due to
seasonal and life cycles during the counting process.
Acknowledgments
This work was promoted by the Scientific Committee for
Oceanic Research (SCOR) which supported the meetings of Working Group 137. The Danish Agency for Science, Technology and
Innovation funded the research program “EXperiences with coastal
eutrophication in a world of human expansion and climate
CHANGE (EXCHANGE)” that funded travel for HHJ. NR acknowledges CSIR-NIO for all the facilitations and the efforts of AAS
Alkawri in collecting and analyzing phytoplankton data of Goa
coast used in this work. AZ's travels to WG 137 meeting was also
partially supported by the EU project Life þ ENVEUROPE and the
MIUR-Flagship Project RITMARE (SP5_WP1). SL's travel to WG 137
meetings was supported by the Academy of Finland (Decision
number 128987), the Finnish Environment Institute (SYKE) and the
Finnish National SCOR Committee. These travel funds allowed us to
meet and discuss the development of this study. Finally, the VELUX
foundation (grant no. VKR022608) supported HHJ to procure microscopes that benefited this project in its initial phase. JS was
supported by the National Natural Science Foundation of China (No.
41276124 and 41176136) and the Program for New Century Excellent Talents in University (No. NCET-12-1065).
References
Almandoz, G.O., Hernando, M.P., Ferreyra, G.A., Schloss, I.R., Ferrario, M.E., 2011.
Seasonal phytoplankton dynamics in extreme southern South America (Beagle
Channel, Argentina). J. Sea Res. 66, 47e57.
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
Q4
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
YECSS4766_proof ■ 16 May 2015 ■ 12/13
12
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
Almany, G.R., De Arruda, M.P., Arthofer, W., Atallah, Z.K., Beissinger, S.R., et al., 2009.
Permanent genetic resources added to molecular ecology resources database 1
May 2009e31 July 2009. Molec. Ecol. Res. 9, 1460e1466.
Boyd, C.M., Johnson, C.W., 1995. Precision of size determination of resistive electronic particle counters. J. Plankton Res. 17, 41e58.
Chan, A., 1978. Comparative physiological study of marine diatoms and dinoflagellates in relation to irradiance and cell size. I. Growth under continuous
light. J. Phycol. 14, 396e492.
Costello, J.C., Chisholm, S.W., 1981. The influence of cell size on the growth rate of
Thalassiosira weissflogii. J. Plankton Res. 3, 415e419.
, M., Dubroca, L., Sarno, D., Zingone, A., Montresor, M.,
D'Alelio, D., Ribera d’Alcala
2010. The time for sex: a biennial life cycle in a marine planktonic diatom.
Limnol. Oceanogr. 55, 106e114.
Davidson, K., Roberts, E.C., Gilpin, L.C., 2002. The relationship between carbon and
biovolume in marine microbial mesocosms under different nutrient regimes.
Eur. J. Phycol. 37, 501e507.
Edlund, M.B., Stoermer, E.F., 1997. Ecology, evolution, and systematic significance of
diatom life histories. J. Phycol. 33, 897e918.
Eppley, R.W., Reid, F.M.H., Strickland, J.D.H., 1970. The ecology of the plankton off La
Jolla, California, in the period April through September 1967. In:
Strickland, J.D.H. (Ed.), Pt III. Estimates of Phytoplankton Crop Size, Growth Rate
and Primary Production. Bull. Scripps Inst. Oceanogr 17, 33e42.
European Commission, 2000. Directive 2000/60/EC of the European parliament and
of the council of 23 October 2000 establishing a framework for community
action in the field of water policy. Official J. Eur. Communities L 327, 1e72.
European Commission, 2008. Directive 2008/56/EC of the European parliament and
of the council of 17 June 2008 establishing a framework for marine Strategy.
Official J. Eur. Communities L 332, 20e44.
European Committee for Standardization (CEN), 2015. Guidance in the Estimation
of Microagal Biovolume (in press). CEN/TC 230 TC230 WI 00230271, CEN/TC
230, Secretariate: DIN.
Gallagher, J.C., 1983. Cell enlargement in Skeletonema costatum (Bacillariophyceae).
J. Phycol. 19, 539e542.
Godhe, A., Kremp, A., Montresor, M., 2014. Genetic and microscopic evidence for
sexual reproduction in the centric diatom Skeletonema marinoi. Protist 165,
401e416.
Harrison, P.J., Conway, H.L., Dugdale, R.C., 1976. Marine diatoms grown in chemostats
under silicate or ammonium limitation. I. Cellular chemical composition and
steady state growth kinetics of Skeletonema costatum. Mar. Biol. 35, 177e186.
Harrison, P.J., Conway, H.L., Holmes, R.W., Davis, C.O., 1977. Marine diatoms grown in
chemostats under silicate or ammonium limitation. III. Cellular chemical
composition and morphology of Chaetoceros debilis, Skeletonema costatum and
Thalassiosira gravida. Mar. Biol. 43, 19e31.
Harrison, P.J., Thompson, P.A., Calderwood, G.A., 1990. Effects of nutrients and light
limitation on the biochemical composition of phytoplankton. J. Appl. Phycol. 2,
45e56.
HELCOM, 2015. Guidelines Concerning Phytoplankton Species Composition,
Abundance and Biomass. Manual for Marine Monitoring in the COMBINE Programme of HELCOM. Annex C-6, Last updated February 2015.
Hillebrand, H., Dürselen, C.-D., Kirschtel, D., Pollingher, U., Zohary, T., 1999. Biovolume calculation for pelagic and benthic microalgae. J. Phycol. 35, 403e424.
Jakobsen, H.H., Carstensen, J., 2011. FlowCAM: sizing cells and understanding the
impact of size distributions on biovolume of planktonic community structure.
Aquat. Microb. Ecol. 65, 75e87.
Jakobsen, H.H., Carstensen, J., Harrison, P.J., Zingone, A., 2015. Estimating time series
phytoplankton carbon biomass: inter-laboratory comparisons of species identifications and comparisons of volume-to-carbon scaling ratios. Estuarine Coast.
Shelf Sci. (this volume).
Jasprica, N., Caric, M., 1997. A comparison of phytoplankton biomass estimators and
their environmental correlates in the Mali Ston Bay (Southern Adriatic).
PSZN1118 Mar. Ecol. 35e50.
Kim, H., Menden-Deuer, S., 2013. Reliability of rapid semi-automated assessment of
plankton abundance, biomass and growth rate estimates: coulter counter versus
light microscope measurements. Limnol. Oceanogr. Methods 11, 382e393.
Koester, J.A., Brawley, S.H., Karp-Boss, L., Mann, D.G., 2007. Sexual reproduction in
the marine centric diatom Ditylum brightwellii (Bacillariophyta). Eur. J. Phycol.
42, 351e366.
Kooistra, W.H., Sarno, C.F.D., Balzano, S., Gu, H., Andersen, R.A., Zingone, A., 2008.
Global diversity and biogeography of Skeletonema species (Bacillariophyta).
Protist 159, 177e193.
Kraberg, A., Baumann, M., Dürselen, C.-D., 2010. In: Wiltshire, K.H., Boersma, M.
(Eds.), Coastal Phytoplankton. Photo Guide for Northern European Seas. Pfeil
Verlag, München, 204.
Leblanc, K., Arístegui, J., Armand, L., Assmy, P., Beker, B., Bode, A., Breton, E.,
Cornet, V., Gibson, J., Gosselin, M.P., Kopczynska, E., Marshall, H., Peloquin, J.,
guiner, B., Schiebel, R., Shipe, R., Stefels, J.,
Piontkovski, S.A., Poulton, A.J., Que
van Leeuwe, M.A., Varela, M., Widdicombe, C., Yallop, M., 2012. A global diatom
database: abundance, biovolume and biomass in the world ocean. Earth Syst.
Sci. Data 4, 149e165.
nchez-Su
Lugo-Vencaino, B.M., Diaz-Ramos, J.R., Sa
arez, I.G., 2003. Biovolumen de
algunas diatomeas centricas de la platforma nororiental de Venezuela. Acta
Cient. Venez. 54, 88e96.
Menden-Deuer, S., Lessard, E.J., 2000. Carbon to volume relationships for dinoflagellates, diatoms, and other protist plankton. Limnol. Oceanogr. 45,
569e579.
Menden-Deuer, S., Lessard, E.J., Satterberg, J., 2001. Effect of preservation on dinoflagellate and diatom cell volume and consequences for carbon biomass predictions. Mar. Ecol. Prog. Ser. 222, 41e50.
Mitra, A., Zaman, S., Ray, S.K., Sinha, S., Banerjee, K., 2012. Inter-relationship between phytoplankton cell volume and aquatic salinity in Indian sundarbans.
Nat. Acad. Sci. Lett. 35, 485e491.
Mizuno, M., Okuba, K., 1985. Seasonal change in the distribution of cell size of
Cocconeis scutellum var. ornata (Bacillariophyceae) in relation to growth and
sexual reproduction. J. Phycol. 21, 547e553.
Moncheva, S., Doncheva, V., Boicenco, L., Sahin, F., Slabalova, N., Culcea, O., 2014.
Report on the MISIS Cruise Intercalibration Exercise: Phytoplankton, ISBN 978606-598-359-5, p. 44.
Moncheva, S., Pantazi, M., Pautova, L., Boicenco, L., Vasiliu, D., Mantzosh, L., 2012.
Black Sea phytoplankton data quality e problems and progress. Turk. J. Fish.
Aquat. Sci. 12, 417e422.
Montagnes, D.J.S., Berges, J.A., Harrison, P.J., Taylor, F.J.R., 1994. Estimating carbon,
nitrogen, protein and chl a from cell volume in marine phytoplankton. Limnol.
Oceanogr. 39, 1044e1060.
Montagnes, D.J.S., Franklin, D.J., 2001. Effect of temperature on diatom volume,
growth rate, and carbon and nitrogen content: reconsidering some paradigms.
Limnol. Oceanogr. 46, 2008e2018.
Mullin, M.M., Sloan, P.R., Eppley, R.W., 1966. Relationship between carbon content,
cell volume, and area in phytoplankton. Limnol. Oceanogr. 11, 307e311.
Nagai, S., Hori, Y., Manabe, T., Imai, I., 1995. Restoration of cell size by vegetative cell
enlargement in Coscinodiscus wailesii (Bacillariophyceae). Phycologia 34,
533e535.
Naz, T., Burhan, Z.U.N., Munir, S., Siddiqui, P.J.A., 2013. Biovolume and biomass of
common diatom species from the coastal waters of Karachi, Pakistan. Pak. J. Bot.
45, 325e328.
€bel, J., et al.,
Olenina, I., Hajdu, S., Andersonn, A., Edler, L., Wasmund, N., Busch, S., Go
2006. Biovolume and Size-classes of Phytoplankton in the Baltic Sea. Baltic Sea
Environmental Proceedings No. 106. HELCOM, p. 144.
Olson, R.J., Vaulot, D., Chisholm, S.W., 1985. Marine phytoplankton distributions
measured using shipboard flow cytometry. Deep-Sea Res. 32, 1273e1280.
Parsons, T.R., Maita, Y., Lalli, C.M., 1984. A Manual of Chemical and Biological
Methods for Seawater Analysis. Pergamon.
Roselli, L., Paparella, F., Stanca, E., Basset, A., 2015. New data-driven method from 3D
confocal microscopy for calculating phytoplankton cell biovolume. J. Microsc.
http://dx.doi.org/10.1111/jmi.12233.
€hl-based phytoRott, E., Salmaso, N., Hoehn, E., 2007. Quality control of Utermo
plankton counting and biomass estimation e an easy task or a Gordian knot?
Hydrobiologia 578, 141e146.
Saravanan, V., Godhe, A., 2010. Genetic heterogeneity and physiological variation
among seasonally separated clones of Skeletonema marinoi (Bacillariophyceae)
in the Gullmar Fjord. Swed. Eur. J. Phycol. 45, 177e190.
Sarno, D., Zingone, A., Saggiomo, V., Carrada, G.C., 1993. Phytoplankton biomass and
species composition in a Mediterranean coastal lagoon. Hydrobiologia 271,
27e40.
Sieracki, C.K., Sieracki, M.E., Yentsch, C.M., 1998. An imaging-flow system for
automated analysis of marine microplankton. Mar. Ecol. Prog. Ser. 168,
285e296.
Stanca, E., Cellamare, M., Basset, A., 2013. Geometric shape as a trait to study
phytoplankton distributions in aquatic ecosystems. Hydrobiologia 701, 99e116.
Strathmann, R.R., 1967. Estimating the organic carbon content of phytoplankton
from cell volume or plasma volume. Limnol. Oceanogr. 12, 411e418.
Sun, J., Liu, D., 2003. Geometric models for calculation cell biovolume and surface
area for phytoplankton. J. Plankton Res. 25, 1331e1346.
Sun, J., Liu, D., Qian, S., 2000. Estimating biomass of phytoplankton in the Jiaozhou
Bay I. phytoplankton biomass estimated from cell volume and plasma volume.
Acta Oceanol. Sin. 19, 97e110.
Taguchi, S., 1976. Relationships between photosynthesis and cell size of marine
diatoms. J. Phycol. 12, 185e189.
Taylor, A.H., Geider, R.J., Gilbert, F.J.H., 1997. Seasonal and latitudinal dependencies
of phytoplankton carbon-to-chlorophyll a ratios: results of a modelling study.
Mar. Ecol. Prog. Ser. 152, 51e66.
Thompson, P.A., Guo, M., Harrison, P.J., 1992. Effects of variation in temperature. I.
On the biochemical composition of eight species of marine phytoplankton.
J. Phycol. 28, 481e488.
Thompson, P.A., Parslow, J.S., Harrison, P.J., 1991. The influence of irradiance on cell
volume and carbon quota for ten species of marine phytoplankton. J. Phycol. 27,
351e360.
Timmermans, K.R., van der Wagt, B., 2010. Variability in cell size, nutrient depletion,
and growth rates of the southern ocean diatom Fragilariopsis kerguelensis
(Bacillariophyceae) after prolonged iron limitation. J. Phycol. 46, 497e506.
Tomas, C.R., 1997. Identifying Marine Phytoplankton. Academic Press, San Diego,
858.
Tsagaraki, T.M., Pitta, P., Frangoulis, C., Petihakis, G., Karakassis, I., 2013. Plankton
response to nutrient enrichment is maximized at intermediate distances from
fish farms. Mar. Ecol. Prog. Ser. 493, 31e42.
Verity, P.G., Robertson, C.Y., Tronzo, C.R., Andrews, M.G., Nelson, J.R., Sieracki, M.E.,
1992. Relationships between cell volume and carbon and nitrogen content of
marine photosynthetic nanoplankton. Limnol. Oceanogr. 37, 1434e1446.
von Dassow, P., Chepurnov, V.A., Armbrust, E.V., 2006. Relationship between growth
rate, cell size, and induction of spermatogenesis in the centric diatom Thalassiosira weissflogii (Bacillariophyta). J. Phycol. 42, 887e899.
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
1
2
3
4
5
6
7
8
YECSS4766_proof ■ 16 May 2015 ■ 13/13
P.J. Harrison et al. / Estuarine, Coastal and Shelf Science xxx (2015) 1e13
von Dassow, P., Montresor, M., 2011. Unveiling the mysteries of phytoplankton life
cycles: patterns and opportunities behind complexity. J. Plankton Res. 33, 3e12.
Waite, A., Harrison, P.J., 1992. Role of sinking and ascent during sexual reproduction
in the marine diatom Ditylum brightwellii. Mar. Ecol. Prog. Ser. 87, 113e122.
Wang, Y., Li, R., Dong, S., Sin, P., Wang, X., 2011. Relationship between cell volume
and cell carbon and nitrogen for ten common dinoflagellates. Acta Ecol. Sin. 31,
6540e6550 (in Chinese with English abstract).
Wheeler, P.A., 1999. Cell geometry revisited: realistic shapes and accurate determination of cell volume and surface area from microscopic measurements.
J. Phycol. 35, 209e210.
13
Zingone, A., Harrison, P.J., Lehtinen, S., McQuatters-Gollop, A., O'Brien, T., Sun, J.,
Jakobsen, H.H., 2015. Increasing the quality, comparability and accessibility of
phytoplankton species composition time series data. Estur. Coast. Shelf Sci. (this
volume).
Zingone, A., Phlips, E.J., Harrison, P.J., 2010. Multiscale variability of twenty-two
coastal phytoplankton time series: a global scale comparison. Estuar. Coasts
33, 224e229.
Zingone, A., Sarno, D., Siano, R., Mario, D., 2011. The importance and distinctiveness
of small-sized phytoplankton in the Magellan Straits. Polar Biol. 34, 1269e1284.
Please cite this article in press as: Harrison, P.J., et al., Cell volumes of marine phytoplankton from globally distributed coastal data sets,
Estuarine, Coastal and Shelf Science (2015), http://dx.doi.org/10.1016/j.ecss.2015.05.026
9
10
11
12
13
14
15
16