Polar Biol DOI 10.1007/s00300-016-1888-z
ORIGINAL PAPER
Microstructural shell strength of the Subantarctic pteropod Limacina helicina antarctica Clara M. H. Teniswood1 • Donna Roberts2 • William R. Howard3 Stephen G. Bray2 • Jodie E. Bradby1
•
Received: 24 March 2015 / Revised: 28 December 2015 / Accepted: 4 January 2016 Ó Springer-Verlag Berlin Heidelberg 2016
Abstract Anthropogenic inputs of CO2 are changing ocean chemistry and will likely affect calcifying marine organisms, particularly aragonite producers such as pteropods. This work seeks to set a benchmark analysis of pteropod shell properties and variability using nanoindentation and electron microscopy to measure the structural and mechanical properties of Subantarctic pteropod shells (Limacina helicina antarctica) collected in 1998 and 2007. The 1998 shells were collected by a sediment trap deployed at 2000 m, 47°S, 142°E, and the 2007 shells were collected using nets from mixed-layer waters in the region (44°– 54°S, 140°–155°E). Transmission electron microscopy revealed that the shells are composed of a polycrystalline structure, and no obvious porosity was visible. The hardness and modulus of the shells were measured using shell cross-section nanoindentation, across various regions of the shell from the inner to outer whorl. No change in mechanical properties was found with respect to the region of the shell cross-section probed. There was no statistically significant difference in the mean modulus or hardness of the shells between the 1998 and 2007 data sets. No major changes in the mechanical properties of these pteropod shells were detected between the 1998 and 2007 data sets, and we discuss the possible biases in the sampling
& Clara M. H. Teniswood
[email protected] 1
Research School of Physics and Engineering, The Australian National University, Canberra, ACT 2601, Australia
2
Antarctic Climate and Ecosystems Cooperative Research Centre, University of Tasmania, Private Bag 80, Hobart, TAS 7001, Australia
3
School of Earth Sciences, University of Melbourne, Parkville, VIC 3010, Australia
techniques in complicating our analysis. However, quantifying the mechanical properties and microstructure of calcified may still provide insights into the responses of calcification to environmental changes, such as ocean acidification. Keywords Nanoindentation Calcifying marine organism Subantarctic
Introduction A number of studies have suggested a shift in marine ecosystems in response to ocean acidification, which is a result of changes in the ocean’s carbonate chemistry due to its uptake of anthropogenic carbon dioxide (e.g. Doney et al. 2009). In particular, attention has been focused on the impacts of ocean acidification on calcium carbonate shellproducing organisms, many of which show reduced calcification in experimental or natural settings in which seawater carbonate saturation has been reduced by CO2 uptake. Impacts have been observed in calcite organisms, such as foraminifera (e.g. Moy et al. 2009; Uthicke et al. 2013), and aragonitic organisms such as corals (e.g. Langdon and Atkinson 2005) and pteropods (e.g. Comeau et al. 2009; Bednarsˇek et al. 2012a, b). Pteropods are important components of Southern Ocean ecosystems, where they reach densities of hundreds to thousands of individuals per m3 (Hunt et al. 2008) and are significant grazers on phytoplankton and smaller mesozooplankton. Thecosomatous (shelled) pteropods produce shells from aragonite. Aragonite is a form of calcium carbonate which is more susceptible to dissolution than calcite (Mucci 1983). Aragonite shell-producers may be vulnerable to ocean acidification as a result of aragonite
123
Polar Biol
undersaturation being crossed at a higher carbonate ion concentration than for calcite (e.g. Orr et al. 2005; McNeil and Matear 2008). By decreasing carbonate mineral saturation, ocean acidification is likely to affect calcifiers in two ways: firstly by reducing their ability to biocalcify and secondly by increasing the conditions under which their calcified structures dissolve. Many studies have suggested that calcium carbonate formation will be reduced in a higher CO2 ocean for a range of organisms including planktonic foraminifera (Spero et al. 1997; Bijma et al. 1999, 2002; Moy et al. 2009), some coccolithophorids (Riebesell et al. 2000; Zondervan et al. 2001), corals (Gattuso et al. 1998; Kleypas et al. 1999; Gattuso and Buddemeier, 2000) and pteropods (Comeau et al. 2009; Lischka and Riebesell 2012; Manno et al. 2012). Laboratory, field and modelling studies on polar pteropods suggest that shell dissolution will occur rapidly as polar oceans become undersaturated with respect to aragonite (Orr et al. 2005; Manno et al. 2007; Comeau et al. 2009; Fabry et al. 2009; Bednarsˇek et al. 2012a, b). Aragonite undersaturation is projected to occur in the Southern Ocean by 2050 (Orr et al. 2005) or even earlier during winter (McNeil and Matear 2008), under high CO2 emissions scenarios. To date, the research on Southern Ocean pteropods has focused on distribution, abundance and shell-weight and shell-flux studies (e.g. Howard et al. 2011; Roberts et al. 2011) with little information available on the structural and mechanical properties. Indeed, given that much of the discussion about the effects of ocean acidification on marine calcifiers centres around the properties of their shells, a detailed characterisation of the structure and mechanical properties could well provide important information about changes in structural integrity. This paper examines the structural and mechanical characteristics of Southern Ocean pteropod shells to gain insight into the baseline properties and variability of these planktonic gastropods’ aragonite shells under modern ocean conditions. Nanoindentation is a common technique for measuring mechanical properties of biomaterials (Doerner and Nix 1986), as it can measure both hardness and Young’s modulus on microstructural scales at site-specific locations. The Young’s modulus of a material is defined as the ratio of the tensile stress over strain in the elastic regime and is a measure of the material’s capacity to deform elastically. Unlike Young’s modulus, hardness is not a fundamental material property but can be described as ‘resistance to permanent deformation’ and is defined as the applied force over the projected contact area of the nanoindentation. The mechanical properties of a range of calcifying marine creatures have been measured using nanoindentation, including one study by the current authors on
123
pteropod shells (Teniswood et al. 2013). Although many calcifying marine creatures form structures that are aragonite-based, their properties do not solely depend on the mineral type used to form the shells. Indeed, details of the structural arrangement, organic component and measurement scale are critical when understanding the mechanical properties of such systems. For instance, nacre, which is the inner iridescent layer of many marine mollusc shells, is also composed primarily of aragonite in a brick-and-mortar-like microstructure (Bruet et al. 2005; Katti et al. 2006; Espinosa et al. 2011). The hardness of nacre has been measured to be 8.7–10.8 GPa and modulus to be 103–114 GPa, using a maximum indentation load of 1000 lN (Bruet et al. 2005). With a higher load of 10,000 lN, Katti et al. (2006) reported a much lower range of values for hardness and modulus of nacre: 1.32–3.1 and 40.95–56.71 GPa, respectively. Scurr and Eichhorn (2006) investigated the mechanical properties of the aragonitic mollusc Ensis siliqua, which has a crossed-lamellar structure interspersed with thin prismatic layers, and measured a mean hardness and modulus of 3.86 ± 0.10 and 82.4 ± 2.7 GPa, respectively, using nanoindentation. The mechanical properties of the tropical pteropod Cavolinia uncinata, which has an interlocked helical nanofibre aragonite structure, have been found to depend on the orientation of the shell cross-section (Zhang et al. 2011). The transverse shell cross-section had a hardness of 5.2 ± 0.4 GPa and modulus of 85.9 ± 2.7 GPa, using a maximum indentation load of 1200–1300 lN. However, parallel to the shell wall, the hardness was similar at 5.6 ± 0.3 GPa but the modulus decreased to 51.5 ± 1.6 GPa (Zhang et al. 2011). The adult shells of Limacina helicina antarctica, the species studied here, have been previously reported to have a measured hardness of *2.3 GPa and modulus of *45 GPa using a 5000 lN maximum indentation load (Teniswood et al. 2013). Adult shells of the common Subantarctic pteropod L. helicina antarctica consist of three aragonite layers: an inner prismatic layer, a crossed-lamellar layer and a thinner outer prismatic layer (Sato-Okoshi et al. 2010). The organic matrix of mollusc shells is estimated to comprise *5 % of the shell weight (Hare and Abelson, 1964); however, to date there has been no known work on the composition of these proteins in L. helicina antarctica pteropods. Moreover, details of the shell structure, composition and mechanical properties are also not well understood. In this work, we use scanning electron microscopy (SEM), nanoindentation and transmission electron microscopy (TEM) to understand the structure–property relations of L. helicina antarctica shells. In addition, hardness and modulus measurements were taken from shells collected in 1998 to compare against measurements from L. helicina
Polar Biol
antarctica shells collected in 2007 with the aim of understanding the variability of shell integrity in these common Subantarctic calcifiers. This estimate of variability may provide a benchmark against which pteropod shell mechanical properties may be compared in a higher-CO2 Southern Ocean.
Materials and methods Sample collection Two sets of shells were analysed in this study: a set collected from a Subantarctic sediment trap deployed in 1997/1998 and a set collected during the Subantarctic Zone Sensitivity to Environmental Change (SAZ-Sense) voyage in 2007. Two different shell collection protocols were used. Although the shells from both should be comparable in terms of mechanical properties testing as both sets of shells are approximately the same size and were collected in the summer months, we later address the possible influence of collection protocols on the measurements. The collection details for both protocols are outlined below. Sediment traps (1998) McLane 21-cup sediment traps were deployed in the central Subantarctic Zone (SAZ) (47°S, 142°E) of the Southern Ocean (Fig. 1) over a time series starting from 1997 to 1998. The shells in this study were collected to the sediment trap moored at 2000 m, which is below the aragonite
saturation horizon (ASH) in the region (currently *1200 m). Sediment trap collection cups were filled with filtered surface seawater from the trap region, which was treated with sodium chloride (5 g L-1) to increase solution density, sodium tetraborate (1 g L-1) as a pH buffer, mercuric chloride (3 g L-1) as a biocide and strontium chloride (0.22 g L-1) to address dissolution of acantharian skeletons (Bray et al. 2000; Trull et al. 2001). The buffered solution is used to ensure that once biogenic carbonate particles have fallen into the cup, no further dissolution will occur. The shells used in this study were collected from 27 January 1998 to 13 February 1998. The pH of the two sediment cups used was 8.36 and 8.44. Further details on the sediment trap methods, deployment times, collection periods and processing methods can be found in Bray et al. (2000) and Trull et al. (2001). SAZ-Sense voyage (2007) The 2007 shells were collected during the SAZ-Sense voyage, which covered a region from 44° to 54°S and 140° to 155°E between 17 January and 20 February 2007 (Fig. 1). Full SAZ-Sense collection site details are found in Howard et al. (2011). Two types of nets were used to collect pteropods during this voyage: a Ringnet and a Rectangular Midwater Trawl (RMT) net (Roe et al. 1980; Pommeranz et al. 1982). The 0.8 m2 Ringnet was deployed vertically from 100 m depth at a speed of 0.5 m s-1. The net was fitted with a 150-lm mesh and a 20-l, 0.3-m wide non-filtering cod-end bucket. RMT deployments comprised 15-min trawls from 150 m depths at speeds of 0.5–1.9 knots. Sample preservation and processing
Fig. 1 SAZ-Sense sample area (blue hatched diamond) in 2007 and sediment trap mooring location (red filled circle) from 1998 in relation to the oceanographic fronts and zones of the Southern Ocean (after Rintoul and Bullister, 1999). STZ Subtropical Zone, STF Subtropical Front, SAZ Subantarctic Zone, SAF Subantarctic Front, PFZ Polar Frontal Zone, PF Polar Front, AZ Antarctic Zone. (Color figure online)
Sediment trap cup samples were stored in buffered solutions at 4 °C until required for analysis. Prior to preparation for mechanical analysis, pteropods were optically identified to species level and rinsed in 100 % ethanol before being stored in fresh 100 % ethanol at 4 °C. Directly after retrieval, the Ringnet and RMT net samples were placed in 100 % ethanol for 24–48 h after which they were washed with and stored in fresh 100 % ethanol at 4 °C. Before being mounted for mechanical analysis, whole shell specimens were removed from ethanol and dried at room temperature in a desiccator for approximately 24 h. Each shell was then individually embedded in epoxy resin and hardener (EpoxySet, Allied). Note that the organic component of the pteropod was not removed (i.e. via bleaching) so that the shell would be as close to its original condition (i.e. composition) as possible. Each sample was polished to reveal a cross-section of the shell wall using a
123
Polar Biol
series of silicon carbide papers followed by a final polish with a silk cloth embedded with diamond paste of grain size 3 lm and, lastly, 1 lm. An image of a polished shell cross-section is shown in Fig. 2.
of interest was removed using a Ga ion beam operating at 30 kV, with decreasing currents at each stage to remove ion beam damaged material. The TEM used was a Philips CM 300 at an accelerating voltage of 300 kV.
Nanoindentation
Statistical data analysis
A schematic of the experimental procedure is shown in Fig. 2. A Hysitron TriboIndenter with a standard pointed (Berkovich) diamond tip was used to determine the hardness and Young’s modulus of the pteropod shells from 1998 to 2007. A maximum indentation load of 5 mN was applied to the shell cross-section over a period of 30 s and was unloaded over 10 s to reduce experimental time. The hardness and modulus of the samples were then calculated using Hysitron software which is based on the Oliver and Pharr (1992) method. Ten individual shells from 1998 were indented in various areas of the shell cross-sections, totalling 518 separate indentations. Similarly, twelve shells from 2007 were indented across similar areas, totalling 556 separate indentations.
Analysis of variance (ANOVA) was used to test for differences in group means within the two sample sets (e.g. area of shell cross-section) as well as to test for differences in means between the sample sets (1998 and 2007). The statistical software R (version 3.0.2) was used for all analyses, and significance for all tests was based on a P value B0.05. Visual inspection of boxplots gave an indication of differences in variance. An F test for variance confirmed that there was no significant difference in variance between the two sample sets in either hardness (P = 0.28) or modulus (P = 0.07), and thus, ANOVA could be used reliably for further analyses.
Results Characterisation Mechanical properties The shells from both sample sets were imaged to determine shell wall thickness and structure using a FEI Quanta 600 MLA environmental scanning electron microscope (ESEM) in low vacuum mode at 10 kV. Each indented shell cross-section was imaged using the FEI ESEM or a Hitachi SU-70 field emission scanning electron microscope (FESEM) at 1.5 kV. No coating was used for ESEM, and a 4-nm platinum coating was applied to samples imaged in the FESEM. To investigate the material under the residual indents, a dual-beam focused ion beam (FIB) system (FEI Helios 600 NanoLab) was used to prepare a *100-nm-thin lamellae for TEM. The surface of the region of interest was protected from ion milling damage by depositing a strip of platinum over the area first using the electron beam and then the ion beam. Material from either side of the region
Fig. 2 Schematic of the experimental procedure for mechanical property measurements using nanoindentation: shell polished to expose a cross-section; the shell is indented with a pointed diamond tip; an applied load verses penetration depth curve from each
123
Nanoindentation was performed in various areas of the shell cross-section, from the centre whorl to outer whorl as shown in the schematic in Fig. 2. Each shell analysed in this study had a diameter of 1–2.0 mm (Figs. 2, 3a). The positioning of indents on the cross-sectioned shell depended on the availability of shell material as some sections of the shell became dislodged during the polishing process. Figure 4 shows a set of nanoindentation data obtained from a single shell (in this instance from a 1998 shell). The plot shows a typical spread of the nanoindentation data measured from a local area (i.e. within the crossed-lamellar layer) of the shell. In both sample sets, the mechanical measurements varied in a similar fashion. Some regions appeared to be relatively ‘soft’, indicated by the indenter tip penetrating *600–700 nm, whereas other regions were
measurement is produced from which the mechanical properties can be calculated. Three different regions on the shell cross-section are indicated in blue: ‘centre’, ‘middle’ and ‘outer’. (Color figure online)
Polar Biol
Fig. 4 Typical nanoindentation load-depth curves from an array of indents in a 1998 pteropod shell
Table 1 Average modulus and hardness of each shell from the 1998 data set and the 2007 data set Shell year, number
Modulus (GPa)
Hardness (GPa)
01
12.2 ± 1.1
0.50 ± 0.05
02
30.7 ± 3.0
2.41 ± 0.20
03 04
55.2 ± 1.8 12.5 ± 1.2
3.34 ± 0.15 0.81 ± 0.10
05
21.3 ± 1.9
2.27 ± 0.28
06
19.2 ± 1.4
1.35 ± 0.13
07
52.5 ± 1.6
3.93 ± 0.15
08
61.8 ± 1.5
4.54 ± 0.14
09
35.4 ± 1.5
3.27 ± 0.15
10
64.2 ± 1.6
4.02 ± 0.14
01
39.3 ± 4.6
2.72 ± 0.29
02
36.8 ± 2.5
2.21 ± 0.22
03
36.7 ± 3.4
2.93 ± 0.26
04
52.9 ± 2.1
2.60 ± 0.20
05
64.7 ± 1.9
4.18 ± 0.15
06
37.4 ± 2.3
1.39 ± 0.18
07
32.4 ± 2.3
0.88 ± 0.09
08 09
48.2 ± 2.2 29.2 ± 2.1
1.76 ± 0.18 1.78 ± 0.22
10
44.2 ± 3.4
2.10 ± 0.38
11
45.6 ± 3.2
1.61 ± 0.26
12
59.4 ± 3.0
2.75 ± 0.19
1998
Fig. 3 a Environmental scanning electron microscopy (ESEM) image of the whole shell (inset) of a Limacina helicina antarctica pteropod collected during the 2007 SAZ-Sense voyage, and the fracture surface of a broken piece of shell. Here, two of the three structural layers are visible: the thick crossed-lamellar layer and the inner prismatic. The boundary between these two layers is indicated by the dashed line. The outer prismatic layer cannot be seen in this image. b Field emission scanning electron microscopy (FESEM) image of a region of the polished cross-section of a 1998 pteropod shell embedded in epoxy
much harder with the tip penetrating only *200 nm. The average hardness and modulus values and the associated standard error for each shell are calculated and are shown in Table 1. The average hardness for individual shells ranged from 0.5 ± 0.05 to 4.54 ± 0.14 GPa, and the average modulus ranged from 12.2 ± 1.1 to 64.7 ± 1.9 GPa. The areas of the shell on which indents were made were compared within each data set. Each individual indent was assigned a position on the shell cross-section: shell ‘centre’, shell ‘middle’ or shell ‘outer’. The ‘centre’ area is defined as the inner most whorl of the shell cross-
2007
section. ‘Outer’ is the area which has direct contact with the external environment (seawater) at the time of collection: the outer whorl. Finally, the ‘middle’ area is between the ‘centre’ and ‘outer’ (Fig. 2). In the 1998 data set, there
123
Polar Biol
were a total of 74 indents in the ‘centre’, 180 in the ‘middle’ and 182 in the ‘outer’ areas. In the 2007 data set, there were 89 indents in the ‘centre’, 103 in the ‘middle’ and 364 in the ‘outer’ areas, with the number of indents in each area dependent on the shell material available in that area. The measured hardness and Young’s modulus from each of the three areas of the shell (‘centre’, ‘middle’, ‘outer’) were compared within each data set using ANOVA. For the 1998 shells, there was no evidence of a statistically significant difference in hardness [F(2,18) = 0.87, P = 0.44] or modulus [F(2,18) = 0.02, P = 0.98] between the three areas. Similarly, for the 2007 shells, there was no statistically significant difference between areas for the modulus [F(2,22) = 1.26, P = 0.30] or hardness [F(2,22) = 0.66, P = 0.53]. The mechanical properties in each area were then compared across the two data sets. Boxplots of the measurements from each area indicated that the range of hardness and modulus measurements in the ‘centre’ and ‘middle’ area were similar between 1998 and 2007, but the range of hardness values in the ‘outer’ area in 2007 appeared less than in 1998. However, there was no statistically significant difference between the mean hardness of the ‘outer’ area of the shells in 1998 and 2007 [F(1,18) = 1.31, P = 0.27]. Likewise, testing the hardness and modulus from the other areas showed no significant difference between 1998 and 2007 (P [ 0.10 in all cases). In order to investigate any correlation between the local shell cross-section morphology and the measured mechanical behaviour, the indents were imaged using ESEM and FESEM. The local morphology was found to vary considerably both in different locations on each shell and across the sample set. This was the case for both sample sets. Some regions appeared ‘rough’ (e.g. Fig. 3b), others appeared ‘smooth’, and some included a combination of both morphologies [see Teniswood et al. (2013) for such images]. No correlation could be established between local shell cross-section morphology and mechanical response in the 2007 shells (Teniswood et al. 2013). This was repeated for the 1998 shell set and similarly no correlation could be established. Characterisation Each shell consists of a single-coiled tube consisting of two to three layers as shown in Fig. 3a. As previously reported, the shell structure was found to consist of up to three layers, a thin inner and outer prismatic layer and a thicker crossed-lamellar central layer (Sato-Okoshi et al. 2010). However, in the shell samples studied here, it was often difficult to identify the inner and outer prismatic layers (Fig. 3a), and thus, our indentation measurements were largely confined to the crossed-lamellar layer. The shell
123
wall thickness of our samples was measured using SEM and ranged from 5 to 30 lm (depending on the region of the shell measured). Figure 3a shows an image of a crosssection of a shell wall that is *15 lm thick, whereas in Fig. 3b the shell wall is *6 lm thick. Both cross-sections are primarily crossed-lamellar in composition. Bright-field TEM images of a lamella through the crosssection of the shell show variations in the surface roughness. Figure 5a shows crystals of various sizes, and the crossed-lamellar layer can be clearly identified. Figure 5b is a selected area diffraction pattern of the region indicated by a dashed circle in Fig. 5a. It reveals a single-crystal pattern which has been indexed as aragonite. The dark-field image highlights the large size variation of these crystals some of which extend several microns from the surface while others extend less than a micron (Fig. 5c). This variation may be part of the explanation of the variability in nanomechanical measurements where the residual depth of the indent varies between 200 and 600–700 nm. This highlights the need to perform a large number of measurements on such samples and to perform the appropriate statistical analysis before the result can be seen as representative of the sample. Another TEM lamella was made through two residual indents (Fig. 5d). Here, it can be seen that the material under the surface of residual indents is homogenous, and in particular, there are no issues with microporosity, voids or cracks. Furthermore, this lamella shows that the indents are spaced far enough apart that no interactions are occurring, suggesting nanoindentation is a valid technique for measuring the bulk properties of the shell material if a large number of indent sites are considered. Comparison of 1998 and 2007 shells The measured mechanical properties of the shells were compared across the 1998 and 2007 data sets. ANOVA analyses of the averaged yearly data revealed a small but statistically significant difference in hardness between shells from 1998 and those from 2007 [F(1,20) = 4.45, P = 0.05], but no difference in modulus [F(1,20) = 0.06, P = 0.81]. At the 95 % confidence interval (Tukey contrasts), hardness decreased at most between 0.01 and 1.92 GPa from 1998 to 2007. Indeed, boxplots of the means by year suggest that there may be greater variance in 1998 than 2007 for both hardness and modulus as shown in Fig. 6. However, plotting mean hardness and modulus for each shell against the number of measurements per shell showed that there was some evidence of sampling bias. This was due to the fact that more measurements were taken from on average harder shells, meaning that some mean values are known to a great precision than others. To account for possible sample bias, weighting (i.e. number of
Polar Biol
Fig. 5 a Bright-field TEM image showing the characteristically uneven surface which is indented, b diffraction pattern from the area indicated by a dashed circle in a, showing a single crystal, c dark-field image of the same cross-section as a showing large crystals, and
d bright-field image of two residual indents and the underlying material of a shell, showing no cracks, voids or porosity. The residual indents are indicated with an arrow. Both cross-sections are from the same shell collected in 1998
measurements per shell) was removed from the analysis, which resulted in no significant difference in shell hardness between years [F(1,20) = 0.67, P = 0.42]. This will be discussed further in the next section.
Discussion Sample collection methods
Fig. 6 Boxplots of the mean modulus and hardness (GPa) of the 1998 shells (518 indentations from ten shells) and 2007 shells (556 indentations from twelve shells). [The bottom and top of the box indicate the first (Q1) and third (Q3) quartiles, respectively, while the band represents the median (Q2). The whiskers represent data greater than Q1 - 1.5 (Q3 - Q1) and less than Q3 ? 1.5 (Q3 - Q1)]
Before discussing the significance of the mechanical property results, it is worth examining the two sample sets in more detail. The two sample sets were collected using different methods: sediment traps at 2000 m depth and trawl nets at 100–150 m depth. The Subantarctic waters at the collection site are supersaturated with respect to aragonite to *1200 m, while the sediment trap used in this study was moored at 2000 m (below the current ASH). The majority of pteropods collected in the sediment traps are thought to fall through the water column as dead animals into the cups. However, pteropods are diurnal migraters who reside and feed in surface waters at night and descend to deeper waters during the day. In addition, Limacina taxa display escape behaviour of downward swimming. Both of these behaviours may mean a small number of pteropods caught by the sediment traps are living ‘swimmers’ (Harbison and Gilmer 1986). It is not possible to distinguish
123
Polar Biol
between the shells collected through settling from the ‘swimmers’, nor is it possible to know whether some individuals had been partly digested or dissolved before reaching the trap. Thus, the shells used in this study from the 1997 sediment traps may well have differing pathways to collection and this may have affected the mechanical integrity of the shells. However, once collected into the trap, it is important to remember that the shells were kept in a buffered solution above current ocean surface pH such that further dissolution would have been unlikely. In contrast to the sediment trap samples, the 2007 SAZSense voyage collected living pteropods into nets near the surface. The vertical distances travelled by pteropods are taxon-specific, but are of the order of 50–200 m for Southern Ocean pteropods (Hunt et al. 2008; Comeau et al. 2012). Live L. helicina antarctica have been caught by nets as deep as 500 m (Hunt et al. 2008). These distances are well above the ASH of 1200 m, and thus, the 2007 shells were unlikely to have been exposed to unsaturated waters nor digested before collection. We assume that the individual shells in this set experienced similar exposure to aragonite supersaturated waters. Despite the differences in the collection systems discussed above, both sets of shells were collected during summer (January–February). Hunt et al. (2008) propose that L. helicina antarctica have a 1-year life cycle with summertime reproduction. Hence, both sets of shells are likely to be from the same stage in the reproduction cycle and are [150 lm in size. Indeed, carbonate concentration is at its annual maximum in summer making it an ideal period in which to collect healthy calcifying shells (McNeil and Matear 2008). However, seasonality of Xaragonite can be large in the Subantarctic Southern Ocean. For example, in the central and eastern Indian Ocean SAZ (41°S and 45°S, respectively), the seasonal amplitude of CO2 is about 50 latm (Borjes et al. 2008), which produces as large a variation in aragonite saturation state in a single year as estimated for decadal trends (e.g. Midorikawa et al. 2012). Similarly, based on nutrient depletion (the major driver of seasonal aragonite saturation variations) we estimate large seasonal variations of Xaragonite, in the SAZ (e.g. Lourey and Trull 2001). Interannual variations in mean-annual airsea pCO2 fluxes are also known to be large in the Southern Ocean, up to *25 % of the overall mean-annual decadal net uptake rate (Lenton et al. 2013), and these variations may affect the comparison between 1998 and 2007 samples. Using nanoindentation to compare sample sets Nanoindentation provides site-specific measurements of mechanical properties at the micro- to nanoscale of a range of materials, including biomaterials. This makes
123
nanoindentation a convenient tool to measure changes in mechanical integrity of hard marine-based biomaterials. An initial study of nanoindentation of L. helicina antarctica pteropods by Teniswood et al. (2013) proposed a standard set of parameters (force, loading and unloading rates) that indicated an accurate measurement of the hardness and Young’s modulus of the pteropod shells. These parameters have been used for both sets of samples in this study. The use of consistent experimental parameters allows us to make meaningful comparisons of nanoindentation data. This consistency of measurement is important, as hardness is not a fundamental material property and is well known to change with varying indentation conditions (Fischer-Cripps 2004). The mechanical properties of the areas (‘centre’, ‘middle’ or ‘outer) on which the shell cross-section was indented were compared. Within each sample set, there was no evidence of differing mechanical properties across the three areas of the shell. This suggests that the mechanical integrity of shell material is consistent through the individual’s life cycle as the shell grows. In addition, there was no difference in the mean mechanical properties in each area of shell between shells from 1998 and 2007. The variability of the measurements were similar in the ‘centre’ and ‘middle’ areas between each set, but was less variable in the ‘outer’ area of the 2007 set compared with the 1998 set. This indicates that the natural variability in ‘outer’ shell strength in 2007 is less than the variability in 1998 shells. Comparing the overall mean hardness between the two sample sets showed some evidence of a statistically significant difference, with the 1998 shells being slightly harder than the 2007 shells. There was no evidence of a difference in modulus. Sample bias in the 1998 data set may have played a role in our results. In the 1998 set, ‘harder’ shells were indented more often than ‘softer’ shells. This could be accounted for by a problem encountered during the sample preparation process, in which it was not uncommon for sections of the shell to become dislodged. This had the effect of limiting the exposed shell cross-section available for nanoindentation measurements. It is likely that the sections dislodged during polishing were ‘weaker’ or fractured, leaving ‘stronger’ areas intact. Thus, the apparent sampling bias led to a probing of more material which could be called ‘hard’. To address this possible bias, the weighting of the number of measurements taken from each shell was removed from the analysis. When the analysis between the two data sets was repeated without weighting, there was no evidence of a difference in hardness. Thus, we are unable to provide conclusive evidence of an effect of time on the hardness between the L. helicina antarctica shells collected in 1998 and those collected in 2007.
Polar Biol
Given concerns over the possible impacts of ocean acidification on calcifiers such as pteropods, would any acidification-driven difference in shell mechanical properties would be detectable over 9 years? The mean rate of summertime pH decrease was 0.0011 to 0.0013 year-1 in the waters north of the Polar Front from 1969 to 2003 (Midorikawa et al., 2012). At this rate, the decrease in pH from 1998 to 2007 would be of the order of 0.01. Although pH is a logarithmic scale, it is not clear that change of this magnitude would be likely to yield significant effects on the mechanical properties of these marine calcifiers. This decadal decrease in pH remains an order of magnitude less than the experimental pH changes yielding significant decreases in pteropod calcification rates (Comeau et al., 2009). However, our study provides a reference point for future measurements of pteropod shells. Furthermore, interannual summer pH decline (0.0020 year-1) is more rapid in the Polar Zone (Midorikawa et al., 2012), suggesting investigation of pteropod shells collected from waters south of the Polar Front would be of significant further interest. Acknowledgments The authors acknowledge Dr. Thomas Trull for the sediment trap samples, which was supported by the Australian Government through the Department of Industry Cooperative Research Centres Program, the Australian Antarctic Sciences program (AAS #1156) and the Australian Marine National Facility. The SAZ-Sense voyage was supported by the Australian Government through the Department of Climate Change, the Australian Cooperative Research Centres Program and the Australian Antarctic Division (AAS Grant #2720). The authors thank Dr. Karsten Goemann and Dr. Sandrin Feig of the Electron Scanning Facility at the Central Science Laboratory, University of Tasmania, for their assistance with the electron microscopy, the ANFF (ACT Node) for use of the FIB, and Dr. Simon Wotherspoon of the Institute for Marine and Antarctic Studies, University of Tasmania, for assistance with the statistical analysis. JEB is funded by an Australian Research Council Future Fellowship.
References Bednarsˇek N, Tarling GA, Bakker DCE, Fielding S, Jones EM, Venables HJ, Ward P, Kuzirian A, Le´ze´ B, Feely RA, Murphy EJ (2012a) Extensive dissolution of live pteropods in the Southern Ocean. Nat Geosci 5:881–885. doi:10.1038/ NGEO1635 Bednarsˇek N, Tarling GA, Bakker DCE, Fielding S, Cohen A, Kuzirian A, McCorkle D, Leze B, Montagna R (2012b) Description and quantification of pteropod shell dissolution: a sensitive bioindicator of ocean acidification. Glob Change Biol 18:2378–2388. doi:10.1111/j.1365-2486.2012.02668.x Bijma J, Spero HJ, Lea DW (1999) Reassessing foraminiferal stable isotope geochemistry: impact of the oceanic carbonate system (experimental results). In: Fisher G, Wefer G (eds) Use of proxies in paleoceanography. Springer, Berlin, pp 489–512 Bijma J, Ho¨nisch B, Zeebe RE (2002) Impact of the ocean carbonate chemistry on living foraminiferal shell weight. Geochem Geophys 3:1064. doi:10.1029/2002GC000388
Borjes AV, Tilbrook B, Metzl N, Lenton A, Delille B (2008) Interannual variability of the carbon dioxide oceanic sink south of Tasmania. Biogeosciences 5:141–151. doi:10.5194/bg-5-1412008 Bray S, Trull T, Manganini S (2000) SAZ project moored sediment traps: results of the 1997–2007 deployments. Antarctic cooperative research centre report no. 15, Hobart, Tasmania, Australia Bruet BJF, Qi HJ, Boyce MC, Panas R, Tai K, Frick L, Ortiz C (2005) Nanoscale morphology and indentation of individual nacre tablets from the gastropod mollusc Trochus niloticus. J Mater Res 20:2400–2419. doi:10.1557/JMR.2005.0273 Comeau S, Gorsky G, Jeffree R, Teyssie JL, Gattuso JP (2009) Impact of ocean acidification on a key Arctic pelagic mollusc (Limacina helicina). Biogeosciences 6:1877–1882. doi:10.5194/bg-6-18772009 Comeau S, Alliouane S, Gattuso JP (2012) Effects of ocean acidifcation on overwintering juvenile Arctic pteropods Limacina helicina. Mar Ecol Prog Ser 456:279–284 Doerner MF, Nix WD (1986) A method for interpreting the data from depth-sensing indentation instruments. J Mater Res 1:601–609. doi:10.1557/JMR.1986.0601 Doney SC, Fabry VJ, Feely RA, Kleypas JA (2009) Ocean acidification: the other CO2 problem. Mar Sci 1:169–192. doi:10.1146/annurev.marine.010908.163834 Espinosa HD, Juster AL, Latourte FJ, Loh OY, Gregoire D, Zavattieri PD (2011) Tablet-level origin of toughening in abalone shells and translation to synthetic composite materials. Nat Commun 2:173. doi:10.1038/ncomms1172 Fabry VJ, McClintock JB, Mathis JT, Grebmeier JM (2009) Ocean acidification at high latitudes: the bellwether. Oceanography 22:160–171. doi:10.5670/oceanog.2009.105 Fischer-Cripps AC (2004) Nanoindentation. Springer, New York Gattuso J-P, Buddemeier R (2000) Calcification and CO2. Nature 407:311–312. doi:10.1038/35030280 Gattuso J-P, Frankignoulle M, Bourge I, Romaine S, Buddemeier R (1998) Effect of calcium carbonate saturation of seawater on coral calcification. Glob Planet Change 18:37–46. doi:10.1016/ S0921-8181(98)00035-6 Harbison GR, Gilmer RW (1986) Effects of animal behaviour on sediment trap collections: implications for the calculations of aragonite fluxes. Deep Sea Res 33:1017–1024. doi:10.1016/ 0198-0149(86)90027-0 Hare PE, Abelson PH (1964) Proteins in mollusk shells. Report of the Director, Geophysics Laboratory, Carnegie Institution, Washington 63:267–270 Howard WR, Roberts D, Moy AD, Lindsay MCM, Hopcroft RR, Trull TW, Bray SG (2011) Distribution, abundance and seasonal flux of pteropods in the Sub-Antarctic Zone. Deep Sea Res II 58:2293–2300. doi:10.1016/j.dsr2.2011.05.031 Hunt BPV, Pakhomov EA, Hosie GW, Sigel V, Ward P, Bernard K (2008) Pteropods in Southern Ocean ecosystems. Prog Oceanogr 78:193–221. doi:10.1016/j.pocean.2008.06.001 Katti KS, Mohanty B, Katti DR (2006) Nanomechanical properties of nacre. J Mater Res 21:1237–1242. doi:10.1557/jmr.2006.0147 Kleypas J, Buddemeier R, Archer D, Gattuso J, Langdon C, Opdyke B (1999) Geochemical consequences of increased atmospheric carbon dioxide on coral reefs. Science 284:118–120. doi:10. 1126/science.284.5411.118 Langdon C, Atkinson MJ (2005) Effect of elevated pCO2 on photosynthesis and calcification of corals and interactions with seasonal change in temperature/irradiance and nutrient enrichment. J Geophys Res Oceans (1978–2012). doi:10.1029/ 2004JC002576 Lenton A, Tilbrook B, Law RM, Bakker D, Doney SC, Gruber N, Ishii M, Hoppema M, Lovenduski NS, Matear RJ, McNeil BI, Metzl N, Mikaloff Fletcher SE, Monteiro PMS, Ro¨denbeck C,
123
Polar Biol Sweeney C, Takahashi T (2013) Sea-air CO2 fluxes in the Southern Ocean for the period 1990–2009. Biogeosciences 10:4037–4054. doi:10.5194/bg-10-4037-2013 Lischka S, Riebesell U (2012) Synergistic effects of ocean acidification and warming on overwintering pteropods in the Arctic. Glob Change Biol 18:3517–3528. doi:10.1111/gcb.12020 Lourey MJ, Trull TW (2001) Seasonal nutrient depletion and carbon export in the Subantarctic and Polar Frontal zones of the Southern Ocean south of Australia. J Geophys Res 106:31463–31487. doi:10.1029/2000JC000287 Manno C, Sandrini S, Tositti L, Accornero A (2007) First stages of degradation of Limacina helicina shells observed above the aragonite chemical lysocline in Terra Nova Bay (Antarctica). J Mar Syst 68:91–102. doi:10.1016/j.jmarsys.2006.11.002 Manno C, Morata N, Primicerio R (2012) Limacina retroversa’s response to combined effects of ocean acidification and sea water freshening. Estuar Coast Shelf Sci 113:163–171. doi:10. 1016/j.ecss.2012.07.019 McNeil BI, Matear RJ (2008) Southern Ocean acidification: a tipping point at 450-ppm atmospheric CO2. Proc Natl Acad Sci 105:18860–18864. doi:10.1073/pnas.0806318105 Midorikawa T, Inoue HY, Ishii M, Sasano D, Kosugi N, Hashida G, Nakaoka S, Suzuki T (2012) Decreasing pH trend estimated from 35-year time series of carbonate parameters in the Pacific sector of the Southern Ocean in summer. Deep Sea Res I 61:131–139. doi:10.1016/j.dsr.2011.12.003 Moy AD, Howard WR, Trull TW, Bray S (2009) Reduced calcification in modern Southern Ocean planktonic foraminifera. Nature Geosci 2:276–280. doi:10.1038/ngeo460 Mucci J (1983) The solubility of calcite and aragonite in seawater at various salinities, temperatures, and one atmosphere total pressure. Am J Sci 283:780–799. doi:10.2475/ajs.283.7.780 Oliver WC, Pharr GM (1992) An improved technique for determining hardness and elastic modulus using load and displacement sensing indentation experiments. J Mater Res 7:1564–1583. doi:10.1557/JMR.1992.1564 Orr JC, Fabry VJ, Aumont O et al (2005) Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature 437:681–686. doi:10.1038/ nature04095 Pommeranz T, Hermann C, Ku¨hn A (1982) Mouth angle of the rectangular midwater trawl (RMT1?8) during paying out and hauling. Meeresforschung 29:267–274 Riebesell U, Zondervan I, Rost B, Tortell PD, Zeebe RE, Morel FMM (2000) Reduced calcification of marine plankton in response to
123
increased atmospheric CO2. Nature 407:364–366. doi:10.1038/ 35030078 Rintoul SR, Bullister JL (1999) A late winter hydrographic section from Tasmania to Antarctica. Deep Sea Res I 46:1417–1454. doi:10.1016/S0967-0637(99)00013-8 Roberts D, Howard WR, Moy AD, Roberts JL, Trull TW, Bray SG, Hopcroft RR (2011) Interannual pteropod variability in sediment traps deployed above and below the aragonite saturation horizon in the Sub-Antarctic Southern Ocean. Polar Biol 31:1739–1750. doi:10.1007/s00300-011-1024-z Roe HSJ, AdeC Baker, Carson RM, Wild R, Shale DM (1980) Behaviour of the Institute of Oceanographic Science’s rectangular midwater trawls: theoretical aspects and experimental observations. Mar Biol 56:247–259. doi:10.1007/BF00645349 Sato-Okoshi W, Okoshi K, Sasaki H, Akiha F (2010) Shell structure of two polar pelagic molluscs, Arctic Limacina helicina and Antarctic Limacina helicina antarctica forma antarctica. Polar Biol 33:1577–1583. doi:10.1007/s00300-010-0849-1 Scurr D, Eichhorn SJ (2006) Analysis of local deformation in indented Ensis Siliqua mollusk shells using Raman spectroscopy. J Mater Res 21:3099–3108. doi:10.1557/jmr.2006.0382 Spero HJ, Bijma J, Lea DW, Bemis BE (1997) Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature 390:497–500. doi:10.1038/37333 Teniswood CMH, Roberts D, Howard WR, Bradby JE (2013) A quantitative assessment of the mechanical strength of the polar pteropod Limacina helicina antarctica shell. ICES J Mar Sci 70:1499–1505. doi:10.1093/icesjms/fst100 Trull T, Bray S, Manganini S, Honjo S, Francois R (2001) Moored sediment trap measurements of carbon export in the SubAntarctic and Polar Frontal zones of the Southern Ocean, south of Australia. J Geophys Res 106:31489–31510. doi:10.1029/ 2000JC000308 Uthicke S, Momigliano P, Fabricius KE (2013) High risk of extinction of benthic foraminifera in this century due to ocean acidification. Sci Rep. doi:10.1038/srep01769 Zhang T, Yurong M, Chen K, Yurong M, Chen K, Kunz M, Tamura N, Qiang M, Xu J, Qi L (2011) Structure and mechanical properties of a pteropod shell consisting of interlocked helical aragonite nanofibers. Angew Chem 123:10545–10549. doi:10. 1002/anie.201103407 Zondervan I, Zeebe RE, Rost B, Riebesell U (2001) Decreasing marine biogenic calcification: a negative feedback on rising atmospheric pCO2. Glob Biogeochem Cycles 15:507–516. doi:10.1029/2000GB001321
