Introduction
The coralline alga Clathromorphum compactum exhibits well-defined
annual growth increments and often produces seasonal conceptacles (seasonal
autumn/winter reproductive structures) within those increments (Adey and
Hayek, 2011; Halfar et al., 2008). The ovoid conceptacles are embedded within
a considerably larger vegetative matrix consisting of a plethora of tissues
(Adey, 1965; Adey et al., 2013) within calcified cell walls. The calcified
cell walls additionally show a variety of well-defined micro-structures in
their high-magnesium calcitic skeleton (Adey et al., 2005; Nash and Adey,
2017, 2018). Owing to C. compactum longevity, multicentennial
chronologies that have the potential to provide data necessary for accurately
calibrating climate models have been constructed (Adey, 1965; Adey et al.,
2015a, b; Halfar et al., 2013). C. compactum grows in subarctic to
high-Arctic oceans in the benthic mid photic zone, and its distribution
becomes severely limited at temperatures above 11–13 ∘C (Adey,
1965).
(a) Cross section of C. compactum mound indicating
major skeletal features. (b) Close-up of meristem,
epithallium, and perithallium cells.
The precise mode of calcification in coralline algae has been long debated
(Adey, 1998; Nash et al., 2018), and the role of photosynthesis in
influencing calcification has not yet been fully established. The shape of Mg
content curves from subarctic/Arctic Clathromorphum sp. suggests
that they continue to grow and calcify for at least part of the winter in
darkness at -1.8 ∘C (Halfar et al., 2013). A recent model of
coralline calcification (Nash and Adey, 2017) shows fibrous cellulosic
extrusions from the cell membranes into the cell wall environment that
provide molecular initiation centres for calcium and magnesium, but it does
not demonstrate a process for cellular injection of Ca and Mg. Their model
indicates Mg control by temperature in a complex subset of calcification
modes but does not include a light component (other than as a basic
photosynthetic requirement).
Production of C. compactum's high-magnesium calcite skeleton occurs
in an intercalary meristem (as in the cambium of higher plants) and is
concurrent with cellular and tissue growth (Fig. 1). The intercalary
meristem, in addition to producing perithallium, the primary body of the
calcified crust, also generates the thin upper layer of the epithallium, the
primary photosynthetic tissue of these plants (Adey, 1965). The perithallial
tissue below the meristem preserves the annual growth increments and the
remains of the yearly conceptacles, which are ovoid conceptacle cavities
(Fig. 1). The primary hypothallium is a multicellular tissue forming the base
of each individual and provides attachment to the substratum (Adey, 1965).
Since it shows a modified form of calcification with higher Mg levels than
the perithallium (Nash and Adey, 2018), it is not utilized in climate
archiving. Wound tissue and secondary hypothallia develop to repair physical
damage, inflicted by wave tools or grazers to the algal thallus (Fig. 1).
While occasionally grazing can damage the meristem and perithallus, most
grazing is restricted to the epithallus. A “symbiotic” association between
chiton and limpet grazing and Clathromorphum sp. has been
demonstrated, and moderate grazing of surficial epithallium is required to
keep the meristem active (Adey, 1973; Steneck, 1982).
Skeletal elemental composition (Mg / Ca) of Clathromorphum sp.
has been shown to correspond to temperature controls (Gamboa et al., 2010;
Halfar et al., 2008; Hetzinger et al., 2018; Williams et al., 2018) and
displays seasonal cyclicity (Adey et al., 2013). However, there is evidence
that light is also influencing Mg incorporation (Moberly, 1968). For example,
in the Newfoundland shallow benthos, C. compactum Mg / Ca ratios
begin to increase in the spring before temperatures increased from the winter
lows (Gamboa et al., 2010). In addition, several studies have noted
inter-sample variability in Mg / Ca (e.g. Chan et al., 2011) and
published Clathromorphum sp. Mg / Ca – temperature calibrations
have been site specific (Hetzinger et al., 2009, 2018; Williams et al.,
2014). It has been hypothesized that, because algae collected at the same site
are experiencing similar temperatures, inter-sample variability may be caused
by differences in shading or orientation relative to the sea surface
(Williams et al., 2014). Differences among calibrations might also result
from collections at different depths or different water clarity between sites
of similar depths.
There is also evidence for the influence of light and temperature on growth
rates of coralline algae. Adey (1970) demonstrated that growth of many
Boreal–subarctic coralline algal genera exhibited a strong relationship with
light at high temperatures and a weak relationship with light at low
temperatures. These patterns suggest growth is limited by photosynthesis in
water temperatures above 4–5 ∘C, while respiration and other growth
processes likely limit growth at lower temperatures (Adey, 1970). Similarly,
Halfar et al. (2011a) found a positive correlation between water temperature
and Clathromorphum sp. growth increment width in the North Atlantic
where winter sea surface temperatures (SSTs) are below 0 ∘C, whereas
Clathromorphum sp. growth in the Bering Sea (winter SST
> 3 ∘C) was unrelated to temperature yet was positively
correlated with light (Halfar et al., 2011b). This supports Adey's (1970)
finding that growth is limited by photosynthetic production in warmer water,
whereas it is temperature-controlled in colder water. Adey et al. (2013)
modelled the relative control of temperature and light in algal systems,
showing that, over a broad range of temperatures and light, temperature had a
somewhat larger effect on productivity than light. Since both temperature and
light are limiting factors, the most limiting will be controlling
productivity. However, in winter subarctic conditions, both factors are at or
near limiting conditions. Similarly, cloud cover, another light proxy, has
been linked to the summer calcification of the rhodolith-forming coralline
alga Lithothamnion glaciale (Burdett et al., 2011). In that study,
lower light levels caused by winter cloud cover reduced summer carbonate
density (Burdett et al., 2011). Additionally, Teichert and Freiwald (2014)
found light to be the most important, and mean annual temperature to be the
second-most-important, physical parameter limiting calcium carbonate
production of coralline algae on the Svalbard shelf. Furthermore, Halfar et al. (2013) used the influence of light on both
growth rates and Mg / Ca to reconstruct sea ice cover in the Arctic. Sea
ice cover constrains growth by limiting photosynthates that the algae produce
(Halfar et al., 2013). Also, bottom temperatures remain relatively constant
below sea ice and more ice-free days allow for higher temperatures, which are
recorded in the Mg / Ca of the algae (Halfar et al., 2013). In summary,
both light and temperature have demonstrated effects on coralline algal
calcification and Mg / Ca.
Contrary to the majority of photosynthetic calcifiers, C. compactum
can thrive in the absence of light for over half of the year. For example,
C. compactum is found in abundance in Arctic Bay, Nunavut, Canada,
at 73∘ N, where sea ice cover causes near darkness at the sea floor
for up to 9 months of the year (Halfar et al., 2013). Regardless of ice,
C. compactum has been found in Novaya Zemlya (Adey et al., 2015b),
an archipelago in the Russian Arctic where there is less than 1 h of
sunlight per day for 4 months of the year. These relationships prevail in
other distantly related coralline genera from high-latitude warmer (Boreal)
waters as shown by the ability of Phymatolithon borealis, P. investiens, P. tenue, and L. glaciale to develop extensive
crusts and mobile rhodoliths in the far north of Norway, where long winter
darkness also occurs (Adey et al., 2018; Teichert et al., 2013). Even though
growth rate is controlled by light and temperature, the chemistry of the
associated calcification does not rely directly on photosynthesis but rather
on total quantity of photosynthates.
Clathromorphum compactum mounds (dark red) up to 3 cm
thick in a typical coralline community at 13 m depth off Quirpon Island in
northernmost Newfoundland. Photo by Nick Caloyianus.
It is clear that a better understanding of the effects of temperature and
light (or lack thereof) on C. compactum growth, calcification, and
elemental composition is necessary to fully understand C. compactum biology and ecology, and the use of this species as a climate
archive. In this study, we examine multiple specimens of C. compactum, monitored at a range of light, temperature, and time treatments in
a suite of tanks having the same open-coast-source water supply. Post-experiment,
multiple samples were analysed for their anatomical and cellular
changes, growth, and MgCO3 composition relative to the various
treatments.
Summer (June) and Winter (December) daily means (dashed curve) and
maxima (black curve) of half-hour-resolution light data from HOBO loggers at
the Bay Bulls collection site. HOBO light sensors were oriented vertically
for maximum light receipt.
Materials and methods
Test subjects
One hundred and twenty-three living specimens of Clathromorphum compactum were chipped off rocky surfaces by divers in August 2012 at
10–12 m depths in Bread and Cheese Cove, Bay Bulls, Newfoundland
(47∘18.57′ N; 52∘46.98′ W). The C. compactum
specimens were dome-shaped, 3–6 cm in diameter, and 1–3 cm thick, a
common size in Newfoundland and Southern Labrador (Fig. 2).
Clathromorphum species are highly distinctive and easily separated
from other coralline species in the region by their surface features and deep
intercalary meristem (Adey, 1965; Adey et al., 2015b). C. circumscriptum, the only other species of the genus in the northwestern
Atlantic, occurs primarily in shallow water, rarely reaching to the depth of
specimen collection in this study, and at maturity is morphologically quite
distinctive (Adey, 1965; Adey et al., 2013). As each collected specimen was
delivered by divers to the dive boat for selection and initial scarring, it
was identified aboard by co-author WHA. During the experiment, each
individual specimen, as selected for analysis, was tagged with a number,
after removal from the tanks, for tracking through analysis; the specimens
are stored in the Smithsonian Institution National Museum of Natural History
(NMNH) with those identification tags for future reference.
Five-day average temperatures through 2011 at Bay Bulls, Bread and
Cheese Cove, at 10 m, taken with HOBO data loggers.
Year-long temperature and light data (lux) were measured with in situ HOBO
loggers (HOBO Pendant; Onset Computer Corporation) at a depth of 12 m at
Bread and Cheese Cove (Figs. 3 and 4). The calibration of the HOBO light
loggers used in the experiment was confirmed against the earlier field data
by installing the same sensors at the field site for a single day. This
provided values similar to the previous time series. In addition,
temperature–depth profiles were obtained from Adey (1966) for four exposed
stations on the east and northeast coast of Newfoundland, along with a
relative abundance–depth profile of C. compactum at those sites
(Supplement Fig. S1). Both sources of instrumental data were used to
establish the temperature and light parameters of the mesocosm complex.
Experimental tank layout at the Ocean Sciences Centre at Logy Bay, NF.
High-light right end of each tank is adjacent to large port-like window and
has fluorescent lights overlying. Opposite, low-light ends of tanks shielded
with opaque plastic sheet. Note cooling probes, coated with ice, placed in 2
and 4 ∘C tanks on left. Three light segments (left to right: low light, ll;
mid-light. ml; high light, hl) with their emplaced specimens can be
seen in right tank.
There were several reasons for not measuring photosynthetically active
radiation (PAR) in this study. (1) Considerable pre-existing field evidence
suggested that calcification in Clathromorphum (and other
corallines) is not directly related to photosynthesis (Nash and Adey, 2018),
but rather only to the availability of stored photosynthate energy. (2) Being
red algae, and having the accessory pigments phycocyanin and phycoerythrin
that supplement chlorophyll, corallines have quite different action spectra
from green algae or higher plants. Also, since these algae reach a peak of
cover at about 20 m, we were working with available light spectra that would
be very different from those at the water surface. Using PAR sensors would
have raised as many questions as it solved. (3) Since C. compactum occurs over a wide depth range (5–30 m), the level of
photosynthesis varies widely, and previous studies have not indicated major
changes in growth and calcification over that range. Thus we would not have
expected a strong relationship between our parameters of interest and light
spectra. (4) Adding PAR, with or instead of lux, especially since action
spectra responses would have been an essential component requiring individual
chambers for each specimen, would have significantly increased the difficulty
and cost of carrying out the experiment.
Experimental setup
The experiment was carried out at the Ocean Sciences Centre (OSC) of Memorial
University of Newfoundland from September 2012 to July 2013. Sea water,
pumped in from a depth of 5 m in the adjacent embayment, Logy Bay, was
provided through a constant-flow system at 1 L min-1 to each tank.
Four 180 L glass tanks were placed so that natural light from large rounded
windows was provided at one end of each pair of tanks, with the opposite ends
of the tanks shaded with black plastic sheets (Fig. 5).
Sixty-centimetre-long, 20 W Hagen Marine-GLO T8 fluorescent tubes were
positioned over the window-lighted end of each tank so as to provide a
significant light gradient (Fig. 6). The light covered one half of each tank
(the high-light section). The immediate darker quarter of each tank was the
mid-light section, and the darkest quarter the low-light section. The
fluorescent tubes were automatically switched on at 10:00 and off at 15:00.
Day length and morning and evening light intensity were supplied by natural
sunlight from the north-facing windows. Experimental temperatures were 2, 4,
7, and 10 ∘C. All tanks were supplied with 4 ∘C seawater
from a master chiller at a constant flow rate of 1 L min-1.
Temperatures in the 7 and 10 ∘C tanks were obtained with immersion
heaters (Hagen, Fluval M300). Temperatures in the 2 ∘C tank were
obtained with two immersion probe coolers (Polyscience, IP 35RCC). September
was a month of gradual temperature change in each tank from roughly
12 ∘C incoming sea water to each experimental value.
Five-day plot of light levels taken from HOBO loggers in
October 2012 in 10 ∘C tank. Pink is high light, blue is medium
light, and green is low light.
(a) Temperature (∘C) and (b) light (lux)
during June in high-light portion of 7 ∘C tank.
HOBO data loggers were placed in each tank at high- and low-light positions to
quantify light and temperature at 5 min intervals throughout the experiment
(Figs. 6–8). A pre-experimental trial with data loggers in all three
sections produced a light value as a proportion of that in the high-light
tank for the remainder of the experiment (Fig. 6). Mean light levels in
October were as follows: high light = 450 lx, medium light = 142 lx, and low
light = 17 lx (Fig. 6). The monthly mean temperatures at each light
level in each tank are shown in Table 1. Occasional changes in flow rates of
the sea water supply as well as heater function required manual system
adjustments to bring temperature to desired values. Due to the limitations of
the laboratory and available equipment, the -1.5 to 1 ∘C
temperature levels representing winter temperatures in coastal Newfoundland
were not achieved, with 2 ∘C being the lowest temperature attained
for the long-term experiment.
Average tank temperatures, November 2012–June 2013.
Tank section
10 ∘C
7 ∘C
4 ∘C
2 ∘C
Nov 2012
High light
10.2
6.9
3.7
2.4
Low light
10.4
7
3.8
2
Average
10.3
6.95
3.75
2.2
Dec 2012
High light
10.1
7.1
4.4
2.2
Low light
10.2
7.1
4.5
1.8
Average
10.15
7.1
4.45
2
Jan 2013
High light
9.3
7
4
2.3
Low light
7
4.4
1.8
Average
9.3
7
4.2
2.05
Feb 2013
High light
10.39
6.95
3.53
2.03
Low light
10.47
6.98
3.65
1.52
Average
10.43
6.97
3.59
1.78
Mar 2013
High light
10.37
7.63
4.12
2.24
Low light
10.48
7.66
4.24
1.67
Average
10.43
7.65
4.18
1.96
Apr 2013
High light
10.05
7.43
4.22
2.13
Low light
10.17
7.45
4.37
1.5
Average
10.11
7.44
4.3
1.82
May 2012
High light
10.91
7.39
4.24
2.31
Low light
10.95
7.46
4.38
1.74
Average
10.93
7.43
4.31
2.03
Jun 2013
High light
10.73
7.09
4.31
2.55
Low light
10.85
7.19
4.42
1.82
Average
10.79
7.14
4.37
2.19
Mean
10.35
7.21
4.14
2
Placement and harvest of C. compactum specimens in
mesocosm
To mark the beginning of the experiment (September 2012), the specimens of
C. compactum were placed in a tank containing approximately 85 mg
alizarin red dye per liter of seawater for 48 h. Alizarin red is
incorporated into the living algal tissue, and it leaves a permanent red stain
line (Kamenos et al., 2008). However, the stain was not incorporated in the
tissues, likely because the test subjects did not grow sufficiently during
the staining process, so staining information was not part of this study.
(a) Temperature (∘C) and (b) light (lux)
during December in high-light portion of 7 ∘C tank.
Each specimen was also laterally scarred (incised) with a fine metal file to
a depth of 200–400 µm aboard the dive skiff immediately following
collection (Fig. 2). The incisions, when sectioned by vertical fracturing,
allowed for a scanning electron microscope (SEM)-based estimate of rate of
wound tissue growth during the experiment, as well as study of the process of
wound repair. Electron microprobe examination was separately applied to
sections of normal (unscarred) perithallial tissue and used to estimate the
beginning of the experiment (shown by the cessation of seasonal Mg
fluctuation). The annual maximum Mg / Ca was used to denote the beginning
of the experiment for electron microprobe data, since this represents highest
temperatures annually at the collection site, which occur in August, at the
time of collection.
The 123 Clathromorphum compactum specimens from Bread and Cheese
Cove were distributed evenly within each of the four experimental tanks, with
∼10 specimens in each light zone (for a total of 30 subjects in each
tank). On the first of each month, beginning with 1 October, individual
specimens were haphazardly collected from each light zone of each tank with
the intention of leaving the remaining specimens evenly distributed over the
zone space so as not to bias in-zone distribution with time. After
collection, specimens were oven-dried for 48 h at 40 ∘C and shipped
to the NMNH for sectioning and anatomical analysis with SEM (Table 2). Using
SEM, the progress of regrowth of the scarred tissue, as well as the status of
the meristem and epithallial and hypothallial tissues, was determined. Of the
123 subjects, 3 were lost to “white-patch disease”, which is
occasionally seen in the wild (Adey et al, 2013), and 85 were vertically
fractured in an attempt both to cut across the scars made immediately after
collection and to section the peak of the mound. The remaining 35 specimens,
mostly collected on 1 July 2013, at the close of the experiment, were also
sent to the NMNH coralline herbarium, as the ClEx Collection, for examination
and voucher storage.
C. compactum growth experiment (ClEx) at Logy Bay, Newfoundland. Specimens
collected in mid-August 2012 and brought to experimental temperature
through September 2012. High light (hl) was < 400 lx, medium light
(ml) was < 160 lx, and low light (ll) was < 17 lx.
Sample date
Temp.
Light
Number
[∘C]
level
harvested
1 Oct 2012
10
hl
1
10
ml
1
7
hl
1
7
ml
1
4
hl
1
4
ml
1
2
hl
1
2
ml
1
1 Nov 2012
10
hl
1
10
ll
1
7
hl
1
7
ll
1
4
hl
1
4
ll
1
2
hl
1
2
ll
1
1 Dec 2012
10
hl
1
10
ml
1
10
ll
1
7
hl
1
7
ml
1
7
ll
1
4
hl
1
4
ml
1
4
ll
1
2
hl
1
2
ml
1
2
ll
1
4 Jan 2013
10
hl
1
10
ml
1
10
ll
1
7
hl
1
7
ml
1
7
ll
1
4
hl
1
4
ml
1
4
ll
1
2
hl
1
2
ml
1
2
ll
1
24 Jan 2013
10
ml
1
10
ll
1
Continued.
Sample date
Temp.
Light
Number
[∘C]
level
harvested
1 Feb 2013
10
hl
1
10
ml
1
7
hl
1
7
ml
1
4
hl
1
4
ml
1
2
hl
1
2
ml
1
1 March 2013
10
dark 1 month
1
10
hl
1
7
dark 1 month
1
7
hl
1
4
dark 1 month
1
4
hl
1
2
dark 1 month
1
2
hl
1
1 Apr 2013
10
hl
1
10
dark 1 month
1
7
hl
1
7
dark 1 month
1
4
hl
1
4
dark 1 month
1
2
hl
1
2
dark 1 month
1
1 May 2013
10
hl
1
10
ml
1
7
hl
1
7
ml
1
4
hl
1
4
ml
1
2
hl
1
2
ml
1
2 Jun 013
10
hl
1
10
dark 1 month
1
7
hl
1
7
dark 1 month
1
4
hl
1
4
dark 1 month
1
2
hl
1
2
dark 1 month
1
1 Jul 2013
10
hl
4
10
ml
1
10
ll
2
10
dark 2 months
1
7
hl
4
7
ml
2
7
ll
3
7
dark 2 months
1
4
hl
4
4
ml
2
4
ll
3
4
dark 2 months
1
2
hl
4
2
ml
2
2
ll
3
2
dark 2 months
1
Anatomy, growth, and calcification in the dark (Adey et al., 2013, 2015) were
also monitored and measured, since C. compactum is primarily an
Arctic species and is seasonally exposed to periods of up to 9 months
without sunlight. On 1 February 2013, and repeated on 1 May 2013,
16 specimens from the low-light areas of each of the experimental mesocosms
were re-scarred (position of scar several millimetres removed from the original scar)
and placed in in situ dark chambers. These specimens were subsequently
collected at 1–2-month intervals (Table 2) and examined in SEM to
determine the extent of wound recovery in the dark (Fig. 9a).
Number of transects analysed with electron microprobe from each
temperature and light level. Number of samples from each level in brackets.
Temperature
High
Medium
Low
(∘C)
light
light
light
10
10 (4)
3 (1)
3 (1)
7
10 (4)
1 (1)
7 (3)
4
10 (4)
6 (2)
7 (3)
2
10 (4)
6 (2)
7 (3)
Each experimental mesocosm had chitons (Tonicella spp.), collected
at Bay Bulls, added in roughly equal numbers to the number of C. compactum mounds present in each tank. The chitons, normally having home
sites on or near C. compactum in the wild, established home sites
beneath the experimental specimens; because the chitons might not travel over
glass from specimen to specimen, we provided one to each C. compactum
mound. The reason for including chitons in the experiment is the
above-described symbiotic relationship between grazing and
Clathromorphum sp.
Sample analyses
All specimens were examined at the NMNH Imaging Laboratory at a range of
magnifications from 50× to 5000× on a Leica Stereoscan 440 SEM
operated at 10 kV, a 13–15 mm working distance, and a sample current of
101 pA. Specimens harvested in July, at the end of the experiment, were
sectioned vertically and polished using diamond suspension to a grit size of
1 µm. The software geo.TS (Olympus Soft Imaging Systems) was
utilized with an automated sampling stage on a reflected-light microscope to
produce two-dimensional maps of the experimental subjects' polished surfaces.
These high-resolution composite images were used to identify the first annual
growth increment and to select linear transects for geochemical analysis
across the annual growth increment, encompassing the length of the experiment
and avoiding the wound area of the samples (for details see Hetzinger et al.,
2009). One to three parallel electron microprobe line transects were analysed
for algal MgCO3 composition on each subject (Table 3) from the
meristem to the first growth line at the University of Göttingen,
Germany, using a JEOL JXA 8900 RL electron microprobe. An acceleration
voltage of 10 kV, a spot diameter of 7 µm, and a beam current of
12 nA were used. Along transects samples were obtained at intervals of
10 µm; to avoid unsuitable areas such as uncalcified cell
interiors, the location of analyses were manually chosen no more than
20 µm laterally from the transect line. Further details of the
method are described in Halfar et al. (2013).
Example of wound recovery of C. compactum. (a)
Section of C. compactum mound through the scar groove, collected
from low-light, 2 ∘C tank on 1 December 2012 after 2.5 months of
recovery. Wound regrowth was initiated with one or more large primordial
cells that gradually transition into normal perithallial cells. Following the
production of three to four typical perithallial cells, meristem cells have
developed and are beginning to produce epithallial cells (ClEx 121 211).
(b) Scar groove section of C. compactum mound from the
high-light, 2 ∘C tank after 6 months of recovery. Groove has nearly
grown in with normal meristem cells and perithallial and epithallial tissue.
See blow-up of large primordial cells in (c).
(c) Primordial cells in lower left of section of (b). These
large ovoid to box-shaped cells typically have massive cell walls that appear
to be a combination of inner cell wall and interfilament, crystals
(ClEx 41 2hl). (d) Partially regrown scar (∼400 µm
at the deepest part) in C. compactum retrieved from 10 ∘C
high-light tank on 1 November 2012. Left side and shallower back portion of
scar have formed new perithallium directly from old, pre-scar tissue.
However, deeper bottom of scar is being filled by new hypothallium being
topped with new perithallium. Allowing for late August and September recovery
without growth, new wound tissue represents 1–2 months' growth. Scale
bars: 100 µm (a, b, d) and 10 µm (c).
Using R version 3.3.2 (R Core Team, 2016) one-way
and two-way analysis of variance (ANOVA) tests were performed to determine
the variance between light and temperature conditions. Assumptions for ANOVA
were met, and normality was verified using the Shapiro–Wilks test, and equal
variance was verified with the Bartlett test. Linear regressions were used to
determine temperature–MgCO3 relationships. There was only data for
three temperature treatments for the medium-light conditions, instead of the
four temperature treatments for high- and low-light conditions. This was
addressed through the degrees of freedom associated with specific ANOVA tests
used to determine each p value.
Examples of wound recovery during dark calcification. (a)
Wound tissue recovery after 2 months in dark conditions. Primordial cells,
perithallial cells, meristem cells and two to three epithallial cells appear
similar to those shown in Fig. 9a–c for lighted wound recovery tissues
(ClEx 71 4dk). (b) Similar wound recovery after 1 month of growth in
the dark. There are fewer cells, with only few meristem cells apparent as
compared to (a); rectangle indicates area magnified in (c)
(ClEx 31 4dk). (c) Magnification section of (b) showing
three meristem and two new epithallial cells being initiated (ClEx 31 4dk).
(d) Section of 1-month dark groove developing new tissues.
Meristem cells are in process of developing along with one or two epithallial
cells. Note clear development of interfilament, even at this early stage,
along with standard radial inner-wall crystals (ClEx 41 4dk). (e)
Normal upper perithallial tissue from 4 ∘C tank after 2 months of
dark growth (meristem cell in upper left) showing extensive, normal diagonal
interfilament crystals developed in the dark (ClEx 71 4dk). (f)
Surface view of part of scar made on 1 May 2013, when specimen was placed in
dark chamber of 4 ∘C tank. Collected on 1 July 2013, large cells
visible on mid-slope lateral surfaces of scar are recovering primordial
cells. Lateral to groove, on normal tissue, right and left, overlying
epithallial tissue can be seen to be heavily grazed by chitons in form of
en-echelon, fine grooves (ClEx 71 4dk). Scale bars: 2 µm
(e), 10 µm (a, c, d), and 100 µm
(b, f).
Discussion
Algal growth characteristics determined from wound recovery
Mean all-temperature, all-light vertical growth of scar (wound) tissue during
the course of this experiment was 42 µm month-1. This would
provide a yearly growth rate of about 500 µm, compared with the
expected mean monthly growth rate of specimens at the Bay Bulls site of
approximately 287 µm yr-1 (24 µm month-1) (Halfar
et al., 2011b). As shown in Fig. 11, total perithallial growth during the
experiment ranged from about 80 to 270 µm (depending on light level)
for an approximate period of 8 months (allowing 1.5 months for
specimen acclimatization), or approximately 120–400 µm per year.
Thus, the growth rate values found in this study are well within the expected
range in the wild. Beginning 6 months into the experiment, fully grown-in
grooves began to be seen. This suggests an even greater differential between
wound recovery rate and expected whole-crust growth rate. Unfortunately, in
this experiment, it was not possible to consistently achieve a tank
temperature below 2 ∘C. Most C. compactum populations are
in localities that reach 0 ∘C or below during the winter. Thus, tank
growth could not be precisely compared with wild growth. Clearly, the
relatively rapid rate of wound recovery is necessary if an even surface on
C. compactum mounds is to be maintained and detritus accumulation
avoided. Such recovery from wounding and conceptacle breakout is frequently
seen in wild-collected specimens (Adey et al., 2013). Chiton grazing is likely
a factor in the delay or absence of scar tissue in some plants. However,
rapid wound recovery demonstrates that temperature and light do not provide
short-term controls on growth and calcification rates.
Growth and calcification are clearly metabolically driven; when a wounded
crust can draw upon photosynthates, stored or in photosynthetic production,
from other parts of the crust, higher rates than normal vegetative growth
under given temperature/light conditions are achievable. Arctic/subarctic
C. compactum crusts must store photosynthate for an extended dark season. The growth rate
that can be accomplished in the short term, to repair damage using stored
energy, is clearly not an option for normal vegetative crust growth. Normal
crust vegetative growth, with energy supplied from epithallial
photosynthesis, must provide not only for local growth but also for annual
reproduction, lateral growth, and wound repair. Thus, the wound repair rates
seen here are higher than the month-to-month growth potential of C. compactum as
controlled by light and temperature and abundantly provided in the
literature (Adey et al., 2015).
Dark growth and calcification
This experiment has demonstrated a strong connection between light and
MgCO3, while Nash and Adey (2017) have shown Mg control only by
temperature. However, in the latter study, light was not a control, other
than as a basic metabolic requirement. Our experiment demonstrated that
growth, with the full complex of cell wall calcification, can occur without
simultaneous light, presumably as long as stored photosynthate is available.
Annual cycles of MgCO3 content of C. compactum
mound from 15 to 20 m in the Kingitok Islands in north-central Labrador.
Cycles show V-shaped pattern, indicating growth halt during winter. Water
temperature at Kingitok is below 0 ∘C for over 200 days per year, and
sea surface is typically frozen from early December to late June (Adey et
al., 2015).
Calcification in the dark indicates that photosynthesis is not the direct
driver of calcification by altering local chemistry. The availability of stored
food previously formed by photosyntheses, growth, and calcification is driven
metabolically. The initiation of the calcite crystal formation cannot be
dependent on photosynthetically elevated pH, as has been proposed in various
studies. C. compactum specimens grown in the wild under Arctic
conditions of 6 months of darkness consistently show a sharp downward spike
in Mg content at the equivalent of 0 to -1.8 ∘C water temperature
(Fig. 14). Full darkness, often under ice cover, occurs before the
temperature reaches its lower limits on the bottom where the plants are
growing. To produce carbonate with MgCO3 ratios equivalent to a
water temperature of -1 ∘C, growth has to have occurred in dark
conditions for some period of time. As we see from this experiment, growth
could proceed for at least 2 months before likely ceasing when sufficient
stored photosynthate is exhausted.
Light and temperature
Results show that both light and temperature significantly affect magnesium
in C. compactum crusts. At lower temperatures (2 ∘C) the
effects of light are relatively small, relating to a 1.4 mol %
MgCO3 (corresponding to 8.75 ∘C on the low-light curve
and 5.4 ∘C on the high-light curve) increase from low to high light.
Differences become larger at higher temperatures (10 ∘C), where
MgCO3 increases by 1.8 mol % (corresponding to
11.25 ∘C on the low-light curve and 6.9 ∘C on the
high-light curve) from low to high light levels. Also, at higher light levels
R2 values indicate a stronger correlation between MgCO3 and
temperature (Fig. 12b). These observations suggest that light and temperature
both result in an increase in MgCO3 and growth, with the effects of
light being more significant at higher temperatures as shown previously for
other coralline species (Adey, 1970).
Implications for proxy
Results show that, although temperature is a major factor contributing to C. compactum MgCO3
incorporation, light must be considered when using C. compactum as a proxy. Our findings
suggest that a global MgCO3–temperature calibration cannot be
produced for C. compactum, because light levels and shading contribute to differences in
MgCO3 within individuals or a large sample. The highest correlation
between temperature and MgCO3 was found when all experimental samples
were combined regardless of light level, suggesting that due to inter- and
intra-sample variability replication is very important when generating
temperature reconstructions, rather than attempting to collect all samples
from similar light conditions. The need for replication caused by
inter-specimen differences has also been highlighted in several studies of
Clathromorphum sp. (Hetzinger et al., 2018; Williams
et al., 2014, 2018).
The effects of light on MgCO3 may explain differences in
Mg / Ca of samples collected from the same site and water depth found in
several C. compactum climate reconstruction studies (Chan et al.,
2011; Williams et al., 2014). Differential shading can be due to temporary
macroalgal overgrowth or the position of the coralline algae, such as under a
ledge or orientation with respect to the surface. In these cases, differing
light levels would necessitate sample-specific calibrations. This would imply
calibrating Mg / Ca for each individual sample with known in situ
temperature before averaging all samples to create a record. Applying this
method could prevent outlier samples that may have experienced significantly
different light levels from having an effect on the proxy reconstruction.
Experiment results also support the use of C. compactum as a sea ice
proxy, because Mg / Ca is driven by both light and temperature. Changes
in light duration on the Arctic seafloor are related to sea ice and thus to
coralline algal Mg / Ca. C. compactum is especially suited as a
sea ice proxy because it is the only high-resolution shallow-marine archive
found in seasonally ice-covered regions of the Arctic, including the
Greenland coast (Jørgensbye and Halfar, 2016), the Canadian Arctic
Archipelago, northern Labrador (Halfar et al., 2013), Novaya Zemlya (Adey et
al., 2015b), and Svalbard (Wisshak et al., 2016).
While the findings of this experiment should inform the use of replication
and calibration of individual samples to improve the use of C. compactum as a climate proxy, they do not discount the results of past
environmental reconstructions using this species. For example, Williams et
al. (2014) used Clathromorphum sp. to reconstruct past temperature
and tested the calibration between Mg / Ca and instrumental temperature
of each sampling location. Also, Halfar et al. (2013) combined annual growth
and Mg / Ca concentrations from C. compactum to reconstruct sea
ice conditions, based on the assumption that these records respond to both
light and temperature. While the influence of both light and temperature on
growth rates of several species of corallines has already been demonstrated
(Adey, 1970), their influences on Mg have now also been confirmed. Based on
our findings, the results of past studies would only be untrustworthy if a
calibration from another study or location was used to convert Mg / Ca to
temperature without confirming this relationship.