Spatio-temporal variation of skeletal Mg-calcite in Antarctic marine calcifiers in a global change scenario

Human driven changes such as increases in oceanic CO2, global warming and pollution may negatively affect the ability of marine calcifiers to build their skeletons/shells, especially in polar regions. Here we address, for the first time, spatio-temporal variability of skeletal Mg-calcite using bryozoan and serpulid species as models in a recruitment experiment of settlement tiles in East Antarctica. Mineralogies were determined using X-ray diffractometry for 754 specimens belonging to six bryozoan species (four cheilostome and two cyclostome species) and two serpulid species from around Casey Station. All species had calcitic skeletons. Intra- and interspecific variability in wt% MgCO3 in calcite among most species contributed to the biggest source of variation overall. Therefore, biological processes seem to be the main factor controlling the skeletal Mg-calcite in these taxa. However, spatial variability found in wt% MgCO3 in calcite could also reflect local impacts such as freshwater input and contaminated sediments. The vulnerability of these species to global change is also examined and those species with high-Mg calcite skeletons and low thermal tolerance (e.g. Beania erecta) could be particularly sensitive to near-future global ocean chemistry changes.

3 Introduction 48 Increases in oceanic CO2, global warming (GW) and pollution will lead to 49 dramatic changes in global ocean chemistry, particularly the carbonate system, in the 50 near future. Some of the expected consequences of the increase in oceanic CO2 are a 51 reduction in seawater pH of 0.3-0.5 pH units by 2100 (ocean acidification; OA), a 52 decrease in the carbonate saturation state (Ω) [1] and metal speciation in seawater [2]. 53 In particular, polar regions are acidifying at a faster rate than elsewhere [3]. 54 Temperature increases will also affect the stability of CaCO3 [4], although the 55 combined effects of OA and GW remain poorly constrained. 56 These human driven changes might negatively affect the ability of marine 57 calcifiers to build their calcified skeletons/shells. Marine calcifiers produce a variety of 58 mineralogical forms (polymorphs) including aragonite, calcite and calcite minerals 59 containing a range of magnesium (Mg) content. Organisms with high-Mg calcite 60 structures are predicted to be more vulnerable to OA as the solubility of calcite increases 61 with its Mg-calcite content [5]. The Mg-calcite in echinoderm skeletons, for example, 62 is expected to increase with GW [6], although this increase may be limited and 63 solubility may not increase dramatically at higher ocean temperatures. Organisms at 64 high latitudes may also be more susceptible to contamination (e.g. trace metals and 65 Persistent Organic Pollutants (POPs)) compared to other regions due to their slower 66 metabolism, growth and larval development and consequent slower detoxification 67 processes, and slower colonisation rates [5,[7][8][9][10]. Synergistic effects of anthropogenic 68 driven environmental change could mean a significant degradation or even loss of 69 calcareous reef habitats and sediment deposits used as shelter, substrate and food for 70 many benthic communities [5,11,12]

Intra-and interspecific variation in skeletal Mg-calcite
188 Mineralogies were determined for 754 specimens belonging to six bryozoan 189 species (four cheilostome and two cyclostome species) and two serpulid polychaete 190 species (Fig 3-6). All species were entirely calcitic.

204
There was a strong significant difference in the mean wt% MgCO3 in calcite 205 among species (PERMANOVA pseudo-F = 226.9, p = 0.001) and sites (pseudo-F = 206 2.04, p = 0.001) (Fig 3-6; Table 2). There was a significant difference among depths 207 (pseudo-F = 5.08, p = 0.029). There were no significant interactions between species, 208 year or depth, indicating differences between species and depth were consistent. There 209 was no effect of year (age of colony). 210 Differences in skeletal Mg-calcite between species contributed to >62% of the 211 total variation observed (Fig 3; Table 2). Variation within species (residual variation 212 among replicates within same depth, year and site) contributed ~24 % to overall 213 variation. Differences among sites (which are confined to comparisons among the same 214 species, depths and years as this was a nested factor in the analysis, i.e. site was nested 215 in species, year and depth) contributed ~9 % of the variance ( Table 2, Fig 4-6). 216 There was a small but significant effect of depth, with differences in skeletal 217 Mg-calcite between different depths contributing ~5 % to the estimated total variation 218 (Fig 4-6). 219 There was no significant overall effect of human impact on skeletal Mg-calcite 220 concentrations (Table 3) nor any significant effect on any individual species (species x 221 impact interaction, Table 3 To our knowledge, this is the first study addressing the spatio-temporal variability of 228 skeletal Mg-calcite using bryozoan and serpulid polychaete species as models in an 229 experiment of settlement tiles in Antarctica. These two taxonomic groups, and some 230 genera and species studied here (e.g. Arachnopusia, Beania, E. antarctica), are 231 common components of hard-substratum sessile assemblages across a wide range of 232 cold latitudes (e.g. South American Region) [42]. These kinds of studies allow regional 233 and latitudinal comparisons of the effects of changes in global ocean chemistry on 234 marine calcifiers. All species had calcitic skeletons, which is consistent with our 235 expectations from previous research on bryozoans and serpulids from high latitudes 236 [28,[43][44][45][46]. This trend is possibly due to low temperatures favouring the deposition of 237 calcite over aragonite as the latter is more susceptible to dissolution in cold waters [45]. 238 239 11

Inter-and intraspecific variability in skeletal Mg-calcite 240
Inter-specific differences in the mean wt% MgCO3 in calcite were the largest 241 source of variation overall. Remarkably, there were significant differences in the 242 skeletal Mg-calcite between all bryozoan species except between E. antarctica and A. 243 decipiens and between the two cyclostome species. There was a wide range of wt% 244 MgCO3 in calcite measurements in the bryozoan and serpulid polychaete species 245 studied here, from 0.  Experimental evidence for this is limited, but the Mediterranean bryozoan Myriapora 334 truncata (Pallas, 1766), which has a wt% MgCO3 in calcite between 8 and 9.5, has been 335 shown to be vulnerable to ocean acidification at a pH of 7.66, with significant loss of 336 skeleton during a short-term experiment [60].

15
There is an increase of Mg content in cheilostome bryozoans towards lower 338 latitudes, which is attributed to the warmer seawater [44]. In a recent study, higher 339 values of Mg in two common co-occurring Mediterranean bryozoans M. truncata and 340 Pentapora fascialis (Pallas, 1766) were also found when the colonies were exposed to 341