Coral Reefs (2014) 33:31–38 DOI 10.1007/s00338-013-1097-y

REPORT

Age and growth of the cold-water scleractinian variabilis and its reef on SW Pacific seamounts

S. J. Fallon • R. E. Thresher • J. Adkins

Received: 8 May 2013 / Accepted: 21 October 2013 / Published online: 16 November 2013 Ó Springer-Verlag Berlin Heidelberg 2013

Abstract Little is known about growth rates of deep- Introduction water reef-forming corals or the rates at which these reefs accumulate. Such information is critical for determining Although scleractinian corals occur commonly and at high the resilience of the reefs to anthropogenic impacts such as diversity at depths below 50 m (Cairns 2007), little is trawling and climate change. We radiocarbon date live- known about their ecology and demography as compared caught and sub-fossil samples of the bioherm-forming coral to their much better studied, more conspicuous and more Solenosmilia variabilis collected from precisely known easily accessed shallow-water relatives (Freiwald and depths and locations by means of a remotely operated Roberts 2005). Recent surveys document extensive cold- vehicle on seamounts south of Tasmania, Australia. The water reefs on seamount and other rocky substrates glob- growth rate of colonies live-caught between 958 and ally, many of which have been or are likely to be subject to 1,454 m, which spans most of the depth range of the spe- the impacts of deep-water fishing (Pitcher et al. 2007; cies locally, ranged from 0.84 to 1.25 mm linear extension Roberts et al. 2009). Age and growth studies carried out on yr-1 and tended to be higher in the deeper-caught material. a few of the deep-water taxa suggest that many are slow- Analysis of skeletal microstructure suggests annual depo- growing and long-lived, usually in excess of a century sition of growth increments near the growing tips, but not (Risk et al. 2002; Andrews et al. 2002; Adkins et al. 2004; closer to the base, as the skeleton is extended and thick- Roark et al. 2005; Sherwood and Edinger 2009), which has ened. Dating of sub-fossil material indicates S. variabilis raised concerns about the resilience of deep-sea coral has been present on Tasmanian seamounts for at least the communities to adverse anthropogenic impacts (Koslow last 47,000 yrs and a reef accumulation rate of et al. 2001; Althaus et al. 2009). 0.27 mm yr-1. The bioherm-forming colonial coral, S. variabilis,is widely distributed on seamounts globally (Davies and Keywords Accumulation rate Annuli Guinotte 2011). In the SW Pacific, it is the dominant reef- Demography Deep sea Growth rate forming that builds extensive reefal structure at a depth band of roughly 1,000–1,300 m on seamounts off New Zealand and Australia (Koslow et al. 2001; Thresher Communicated by Geology Editor Prof. Bernhard Riegl et al. in press). These reefs are relatively high in species S. J. Fallon richness, suggesting that they are important habitat for Research School of Earth Sciences, Australian National seamount biota. They are also fragile and easily damaged University, Canberra, ACT, Australia by near-bottom trawling (Althaus et al. 2009) and may be R. E. Thresher (&) particularly susceptible to adverse effects of climate CSIRO Climate Adaptation Flagship, GPO Box 1538, Hobart, change in general and ocean acidification in particular TAS 7001, Australia (Guinotte et al. 2006). To an extent, this apparent vulner- e-mail: [email protected] ability depends on the growth rate of the coral and its J. Adkins ability to regenerate after damage and to grow into and California Institute of Technology, Pasadena, CA, USA colonize new habitat. To begin to quantify this resiliency, 123 32 Coral Reefs (2014) 33:31–38 we use radiocarbon dating to determine the ages and intermediate positions of three live-caught colonies. growth rates of S. variabilis collected at precisely known Specimen 1 was 10.7 cm long and was live-caught at depths and locations off SW Pacific seamounts by means of 958 m on North Sisters seamount, at 44°16.60S., the remotely operated vehicle Jason2 (hereafter Jason). We 147°15.60E. Specimens 2 and 3 were 11.0 and also dated sub-fossil material to provide, for the first time 8.7 cm long, respectively. They were collected on globally, an estimate of the accumulation rate of a very seamount K1 (44°17.60S., 147°23.20E) at 1,238 and deep-water scleractinian reef. 1,454 m depth, respectively. By design, the specimens were chosen as to span most of the depth range of the coral on the Tasmanian seamounts. Specimen 1 was Methods collected just below the shallowest live specimen seen on the Jason dive program (at 890 m); specimen 2 Field work was carried out and samples obtained from the coincided with the depth of peak reef development ‘‘Southern Seamount’’ region south of Tasmania, Australia. (1,050–1,275 m); and specimen 3 was the deepest live The area is described in detail and seamount locations S. variabilis seen during the Jason program. given by Koslow et al. (2001). The seamounts are of vol- 2. As a first attempt to determine the accumulation rate of canic origin, come to within about 750 m of the surface, the reef as a whole, we dated samples collected from a and support an extensive benthic community, dominated at large cut in the reef that we found near the peak of depths \ 1,300 m by the reef-forming scleractinians, S. seamount K1, at 1,231 m depth (44°17.5660S., variabilis and Enallopsammia rostrata, and deeper by 147°23.2120E). The ROV pilot estimated the cut to hormathiid anemones, bathylasmatid barnacles and isidid be about 1.67 m deep and about twice that wide. One gorgonians (Koslow et al. 2001; Althaus et al. 2009; edge was sheer and vertical, the other more gradual, Thresher et al. 2011a). All samples were obtained from a with the cut itself oriented across the top of the reef. survey of the seamounts in December 2008/January 2009 From discussions with local researchers and fishers, it by the Woods Hole Oceanographic Institution ROV Jason is likely the cut dated from 1997, when CSIRO (for details, see Thresher et al. in press). Samples were targeted the top of K1 for biological sampling (see taken using a small basket scoop at the end of one of the Koslow et al. 2001). To assess the accumulation rate of ROV’s manipulator arms. Samples were sorted immedi- the reef, we took two samples from the cut. One was ately upon return to the surface and frozen at -70 °C. made by digging the mesh basket into the approximate The gross skeletal morphology of S. variabilis consists mid-point of the vertical wall of the cut, and the second of a tubular structure with irregularly spaced but frequent, by brushing away loose fragments and then digging the three-dimensional dichotomous branching. In most basket into the bottom of the cut. We also selected for instances, one branch bears a terminal polyp, whereas the dating several well eroded and, we suspected, old other continues developing as the main extension axis. samples collected opportunistically elsewhere on the Samples for radiocarbon dating, increment counts and shallower (\1,400 m) seamount complex to begin to measurements of skeletal dimensions were all taken from gauge the maximum age of the S. variabilis reef the extension axis, at points of relatively uniform diameter locally. just below where the skeleton flares into the next branch- The method for specimen handling and analysis was ing. Sampling was done using a Dremel tool, cutting the similar for both sample sets. Each coral sample was sub- coral perpendicular to the extension axis. Skeletal diameter jected to physical pretreatments. The outer surfaces were was measured using a digital caliper. ground off with a dental drill to expose clean white calcium Extension (growth) rates were calculated using geomet- carbonate. The inside was also drilled out to remove the ric mean regressions of the distance from the tip of each interior dissepiments, resulting in a hollow tube. The hol- coral, as measured along the bifurcating extension axis, low tube was coarsely crushed into *0.25-mm chunks and against the mean calendar year age estimate for each sam- subjected to a 0.1 M HCl leach in which *50 % of the ple. Statistical procedures follow Sokal and Rohlf (1981). outside material was removed. The resulting clean car- bonate was rinsed 5 9 in 18 MX water (Milli-QTM) until Radiocarbon sampling and analysis near neutral pH was achieved. A further subsample of 8–10 mg of CaCO was loaded Two sets of S. variabilis samples were dated using radio- 3 into a 10 ml serum vial (BD VacutainersÒ) and evacuated, carbon analysis. before CO2 was liberated by the introduction of 85 % 1. To assay extension rates of individual colonies, orthophosphoric acid (Ajax UNIVAR, Analytical Grade). subsamples were taken from the base, tip and The CO2 was then passed through a cryogenic water trap,

123 Coral Reefs (2014) 33:31–38 33

of overgrowth on the radiocarbon age (Fig. 1; see also Adkins et al. 2000).

Increment analysis

Possible incremental growth structure was examined in Specimen 1. Extended sections of the coral were brushed clean of adhering dried organic matter, and then embedded in epoxy (Araldite Kit K3600). Approximately 200 lm sections were cut using a Steuers Accutom saw, which were then attached to glass slides using epoxy. Each sec- tion was hand ground using a range of fine sandpapers to a thickness of about 25 lm, before being polished using 3-lm grit. Structure was observed and increments counted using a Leica MX16FA dissecting microscope and a Leica DMIRB inverted microscope. Photographs were taken Fig. 1 Radiocarbon age of single sample after collecting the first using the Zeiss Axiocam system. 25 %, then successive 25 % of CO2 evolved from sample measured for % carbon yield and transferred to an indi- Results vidual graphite reactor assembly. In the presence of hydrogen and using Fe powder as a catalyst and a tem- Age and growth perature of 570 °C, the CO2 is converted to graphite, and the resulting water from the reaction is trapped using Mg Calculated instrumental measurement error (1 sigma) for perchlorate. The graphite is then loaded into Al cathode each age estimate was approximately ±45–55 yrs, but sample holders for 14,13,12C isotope analysis on the single- replicate analysis of two subsamples of the same segments stage accelerator mass spectrometer located at the Research near the base and tip of specimen 2 differ by only 18 and School of Earth Sciences, The Australian National Uni- 8 yrs, respectively (Fig. 2b), providing an empirical mea- versity (Fallon et al. 2010). All samples were normalized to sure of the precision of the age estimates. Radiocarbon Oxalic Acid I (Stuvier and Pollach 1977) and background dating of five sections from specimen 1 indicates a colony 14 subtracted using C free CaCO3. Conversion of radiocar- age of approximately 120 yrs and an average linear bon years to calendar years was done by subtracting 14C extension rate of *0.85 mm yr-1 (95 % CI reservoir age (estimated from the outermost sample for 0.13–1.56 mm yr-1) (Fig. 2a). Ages and growth rates for each specimen) and preformed using OXCAL 4.1 (http:// specimens 2 and 3 are approximately 85 yrs and c14.arch.ox.ac.uk/oxcal/). The quoted uncertainty on the *1.2 mm yr-1 (95 % CI 0.48–1.96 mm yr-1), based on radiocarbon age (1 sigma) follows Stuvier and Pollach six analyses of four sections, and 75 yrs and (1977). *1.25 mm yr-1 (95 % CI 1.0–1.48 mm yr-1), based on analysis of five sections, respectively (Fig. 2b, c). Leaching experiment Increment structure In order to test whether our results were being influenced by any outer coatings on the samples, we prepared a Thin sectioning of specimen 1 (locations are shown in larger sample and only mechanically treated the outside Fig. 3a) revealed an internal structure that consists of and inside of the sample as explained above. An aliquot inward projecting septa surrounded by a calcium carbonate of sample (*35 mg) was loaded into the glass septa wall that was spanned by prominent radially oriented sealed vial, and enough orthophosphoric acid was added crystal bundles. Both the septa and the skeletal wall had a to evolve *25 % of the CaCO3 sample; this was repeated fine incremental structure, the former suggestive of an in- 3 additional times so that we had four samples, the filling and thickening process and the latter skeletal samples containing 0–25, 25–50, 50–75 and 75–100 % of thickening by deposition of annuli. The annuli were the material. This successive leaching technique has been extremely low contrast, difficult to visualize, photograph or used in 14C studies of surface-exposed corals to assess the count and had an average width of about 200 lm (Fig. 3b, effect of overgrowth contamination (Yokoyama et al. c). The number of annuli increased non-linearly from the 2000). Our results suggest that there is little-to-no effect tip to the base of the specimen (Fig. 3d), ranging from 5 to 123 34 Coral Reefs (2014) 33:31–38

Fig. 2 a–c Length at age plots a for three live-caught Southern d Hills Solenosmilia variabilis. Extension rates are slopes and 95 % confidence intervals of geometric mean (type 2) regressions between distance and age. d Relationship between collection depth and mean estimated extension rates. Solid circles are for New Zealand samples as reported by Neil (pers comm), for which depths b are the mean of the benthic trawl from which the samples were taken. Open circles are the Australian Southern Hills samples

c

11 (range of counts) in a section cut 8 mm from the tip of cut, the regression between depth into the cut and the age the coral to 65–70 near the base. Longitudinal differences of the coral fragments is linear (R2 = 0.98) and has a slope in the number of counted annuli correlated with changes in of 0.27 mm yr-1 (95 % CI 0.23–0.31 mm yr-1) (Fig. 4b). the diameter of the skeleton, both increasing from the tip to The oldest S. variabilis fragment had a radiocarbon age about 60 mm from the tip and stabilizing thereafter of 47,400 ± 800 yrs BP, collected at 1,500 m and uncor- (Fig. 3d). rected for a reservoir age. The modern radiocarbon reser- voir age at 1,500 m is about 1,300 yrs (Lassey et al. 1990). Reef age and accumulation rate

The analyzed sub-fossil fragments from the cut on K1 were Discussion covered with a ferro-manganese coating, but were other- wise intact and showed little evidence of carbonate dia- The extension rates of the live-caught S. variabilis we genesis (Fig. 4a). The mean estimated ages (radiocarbon, examined are similar to those reported for a live-caught and model-adjusted) of three samples collected at the base of two recent sub-fossil samples in New Zealand by Neil et al. -1 the cut ranged from 6,236 to 6,851 cal BP, averaging (pers comm) (0.25–1.56 mm yr ) and to the only other 6,624 cal BP (minimum and maximum values across all reported in situ growth rate for a deep-water bioherm- -1 specimens, 6,075 and 6,974 cal BP, respectively). This forming scleractinian (*5mmyr for a North Atlantic assumes a reservoir age of 1,250 yrs, the approximate specimen of E. rostrata, by Adkins et al. 2004). More work modern value at that depth based on the marginal values of has been done on cold-water species in shallower waters. In the live corals we analyzed. Mean estimated ages for three situ tagging experiments on transplanted fragments of samples extracted from the approximate mid-point of the Lophelia pertusa at a relatively deep site (430–520 m) in cut ranged from 3,530 to 3,798 cal BP, averaging 3,638 cal the Gulf of Mexico indicate extension rates of -1 BP (minimum and maximum values ranging from 3,377 to 2.44–3.77 mm yr (Brooke and Young 2009), with higher -1 3,798 cal BP). Assuming an age of 0 yr BP at the top of the rates (up to 25 mm yr ) reported for the species in 123 Coral Reefs (2014) 33:31–38 35

Fig. 4 a Representative sample (Sv 25) of sub-fossil Solenosmilia variabilis fragment collected by ROV from the base of the trawl cut in K1, with an estimated calendar year age of 6,851 (±123) years BP. b Depth profile of calendar ages of sub-fossil S. variabilis fragments collected from the base and approximate mid-point of the trawl cut in K1

study suggested an average extension rate for the species of about 5 mm yr-1 (Orejas et al. 2011). In situ and labora- tory studies of shallow-water specimens of the solitary Fig. 3 Solenosmilia variabilis specimen 1. a Live collected specimen cold-water scleractinian Desmophyllum dianthus indicate indicating radiometric age determination (paired solid bars), sections linear extension rates of *2–5 mm yr-1 (Jantzen et al. for annuli counts (dashed lines) and locations of sections shown in images b and c. b, c Transmitted light microphotographs of sections 2013). Radiometric dating of a deep-water specimen (Ad- taken 8 mm from the terminal polyp and 5 mm from the fragment kins et al. 2004) and modal analyses of size–frequency base. d Skeletal diameter (thick bars, indicating minimum and distributions of SW Pacific samples (Thresher et al. 2011b), maximum) and number of counted skeletal annuli (fine bars)asa however, suggest extension rates similar to those of S. function of position along the extension axis. Capped lines indicate -1 the range of counts in each section. e Calendar calibrated radiocarbon variabilis (\2mmyr ) for the species at depths similar to age; regression indicates an age of *95 yrs and extension rate of those we examined. *0.85 mm yr-1. Arrows indicate age range estimates, curved lines In the SW Pacific D. dianthus, mean size increases with 95 % CI of regression line depth, peaking at about 2,100 m (Thresher et al. 2011b), which is reminiscent of the higher extension rates with shallower water and in aquaria (Roberts et al. 2009). Rel- increasing depth across the six S. variabilis specimens aged atively high extension rates (up to 18 mm yr-1) have also to date (Fig. 2d). Why apparent growth rates should be been reported for Mediterranean Sea specimens of Mad- higher in deeper-water specimens is not clear, particularly repora oculata held in relatively warm aquaria as the latter are below the modern aragonite saturation (11.5–12.5 °C) (Orejas et al. 2008), though a subsequent depth (ca. 1,050 m; see Thresher et al. 2011c) and hence 123 36 Coral Reefs (2014) 33:31–38 nominally expending more resources to maintain skeletal growth and integrity, are in increasingly colder water, and are farther from the presumed surface food source. These observations, however, would benefit from validation using alternative dating methods. Our age and growth estimates assume a radiocarbon reservoir age that is constant at depth. Although our estimates are broadly consistent with the literature (and with counts of annuli in specimen 1), changing reservoir ages as a result of decadal scale heave of ocean water masses has the potential to cause a signif- icant over or under-estimation of short-term growth rates. Our preliminary efforts to validate the ages using U/Th ratios were ambiguous due to apparently high levels of Fig. 5 Diagram of the proposed skeletal growth of Solenosmilia exogenously sourced thorium, but alternative methods variabilis. Each year, a new layer of aragonite is deposited over the existing skeleton near the growing tip, resulting in a thickening and (e.g., Pb/Ra ratios) might test both the S. variabilis growth strengthening of the linear growth form. This process is maintained rates and short-term fluctuations in reservoir age. for a period of several decades, after which further thickening ceases. The internal structure for the live-caught S. variabilis is As a result, the number of growth increments close to the growing tip consistent with a ca. 1 mm extension rate per year and at reflects annual ages, whereas those further along the extension axis indicate the age at which thickening stopped least approximately annual band formation. Annual band- ing has previously been suggested for several deep-sea octocorals (Thresher et al. 2004; Sherwood et al. 2005; extension rate of *1mmyr-1, i.e., similar to that indi- though see also Tracey et al. 2007). Among scleractinians, cated by the radiocarbon analysis. Annually periodic internal banding (annuli) has been reported in a number of deposition is consistent with pronounced seasonal cycles of cold-water taxa (e.g., Nagelkerken et al. 1997), but also water temperature in the S. variabilis depth range, on the reported as difficult to visualize and of possibly variable order of 1 °C (Thresher et al. 2010), and surface produc- provenance (Smith et al. 2002). Among deep-sea reef- tivity (Clementson et al. 1998). Thickening of the skeleton, forming taxa, Adkins et al. (2004) present evidence that the however, stops at a distance about 50–60 mm from the annuli in a North Atlantic specimen of E. rostrata form at growing tip, beyond which both the number of counted roughly decadal intervals. The annuli in the skeleton of S. increments and skeletal diameter are stationary (Fig. 5). variabilis, however, are much finer (*200 lm wide), very This proposed growth framework provides a logical basis faint and similar to those reported by Mortensen and Rapp for colony growth, but also implies that annually deposited (1998) in the shallower, cold-water branching species L. increments in this deep-water species are an unreliable pertusa. In the latter, high-density bands appear to be means of aging specimens beyond a certain point, even if annual, among which were a number of fine growth lines the logistical difficulties of working with such low reso- that were suggested to form on a lunar cycle. In S. varia- lution annuli could be resolved. bilis, the number of annuli was often difficult to ascertain, Dating of the older material allows us to provide a first but increased from less than 10 close to the growing tip of ever estimate of accumulation rates for a hermatypic coral the colony to about 60–70 at a distance 50–60 mm from the assemblage at depths greater than a kilometer. Based on the tip and remained at about that level to the base of the ages of the samples collected at the mid-point and base of the 155-mm-long sample. At the same time, the diameter of the trawl cut at the top of seamount K1, we estimate an accu- colony, all of which was covered with live tissue, increased mulation rate of about 0.27 mm yr-1, which corresponds to from about 4 mm near the tip (excluding the flared calyx about 1 meter every 3.1K yrs. Accumulation rates have been around the terminal polyps) to just over 5 mm about estimated for shallower hermatypic mounds for two North 50–60 mm from the tip and remaining about the same Atlantic cold-water scleractinians, L. pertusa and Oculina thereafter to the base of the sample. We interpret these vericosa, based on a combination of dating and acoustic observations as follows: Initial extension of the colony is estimates of reef/mound thickness (reviewed by Roberts relatively rapid as a new terminal polyp forms. Thereafter, et al. 2009). These estimates are orders of magnitude higher the colony increases both in length (further extension) and than we calculated for the South Pacific S. variabilis reef. skeletal width, as new aragonite matrix is deposited on the This could reflect regional differences in productivity, in skeleton, presumably strengthening it. This deposition, we water temperatures or absolute species-specific differences suggest, is annual, up to a point. This is suggested by the in growth characteristics. Modern temperatures for the North rough correspondence between distance from the terminal Atlantic sites are 3–5 °C higher than the 3–4 °C typical of tip and the number of counted increments, with an implied the S. variabilis sites. The slow accumulation rate for the 123 Coral Reefs (2014) 33:31–38 37

S. variabilis reef is also consistent with qualitative obser- distribution of deep-sea scleractinian corals. Front Ecol Environ vations of thick reef complexes (we estimate reef mounds off 4:141–146 Jantzen C, Laudien J, Sokol S, Forsterra G, Haussermann V, Kupprat NE Tasmania at 2–3 m thick, for example) and the apparent F, Richter C (2013) In situ short-term growth rates of a cold- longevity of the reef. Our oldest sample dated thus far for the water coral. Mar Freshw Res 64:631–641 reef complex is just under 50K yrs BP. This age estimate is Koslow JA, Gowlett-Holmes K, Lowry JK, O’Hara T, Poore GCB, near the limit of radiocarbon dating (see Thiagarajan et al. Williams A (2001) Seamount benthic macrofauna off southern Tasmania: community structure and impacts of trawling. Mar 2013) and hence needs to be used cautiously, but suggests the Ecol Prog Ser 213:111–125 seamounts have been occupied by S. variabilis since before Lassey KR, Manning MR, Sparks RJ, Wallace G (1990) Radiocarbon the Last Glacial Maximum (see also Thiagarajan et al. in the sub-tropical convergence east of Tasmania—an interim (2013), for a parallel analysis of a South Pacific solitary report. DSIR Physical Sciences Report #11, DSIR Physical Sciences, Nuclear Sciences Group scleractinian, D. dianthus). Mortensen PB, Rapp HT (1998) Oxygen- and carbon isotope ratios related to growth line patterns in skeletons of Lophelia pertusa Acknowledgments We thank E. Anagnostou, A. Beck, W. Cho, A. (L) (: Scleractinia): Implications for determination of Gagnon, K. Gowlett-Holmes, S. John, A. Kennedy, H. Kippo, N. linear extension rates. Sarsia 83:433–446 Meckler, D. Mills, N. Thiagarajan, D. Staples, and A. Subhas, and in Nagelkerken I, Buchan K, Smith GW, Bonair K, Bush P, Garzo´n- particular, the crews of the ROV Jason and its support vessel RV Ferreira J, Botero L, Gayle P, Harvell CD, Heberer C, Kim K, Thomas T. Thompson for their professional assistance in the field. We Petrovic C, Pors L, Yoshioka P (1997) Widespread disease in also thank Rachel Wood for help with coral dating. Components of Caribbean sea fans: II. Patterns of infection and tissue loss. Mar this work were supported by the National Science Foundation, the Ecol Prog Ser 160:255–263 Australian Department of Environment, Water, Heritage, and the Orejas C, Gori A, Gili JM (2008) Growth rates of live Lophelia Arts, the Australian Commonwealth Environmental Research Fund, pertusa and Madrepora oculata from the Mediterranean Sea and Australian National Climate Adaptations Research Program. maintained in aquaria. Coral Reefs 27:255 Orejas C, Ferrier-Pages C, Reynaud S, Gori A, Beraud E, Tsounis G, Allemand D, Gili JM (2011) Long-term growth rates of four Mediterranean cold-water coral species maintained in aquaria. References Mar Ecol Prog Ser 429:57–65 Pitcher TJ, Morato T, Hart PJB, Clark MR, Haggan N, Santos RS Adkins JF, Henderson GM, Wang S-L, O0Shea S, Mokadem F (2004) (2007) Seamounts: ecology, fisheries & conservation. Blackwell, Growth rates of the deep-sea scleractinia Desmophyllum crista- Oxford, UK galli and Enallopsammia rostrata. Earth Planet Sci Lett Risk MJ, Heikoop JM, Snow MG, Beukens R (2002) Lifespans and 227:481–490 growth patterns of two deep-sea corals: Primnoa resedaeformis Adkins JF, Griffin S, Kashgarian M, Cheng H, Druffel ERM, Boyle and Desmophyllum cristagalli. Hydrobiologia 471:125–131 EA, Edwards RL, Shen C-C (2000) Radiocarbon dating of deep- Roark EB, Guilderson TP, Flood-Page S, Dunbar RB, Ingram BL, sea corals. Radiocarbon 44:567–580 Fallon SJ, McCulloch M (2005) Radiocarbon-based ages and Althaus F, Williams A, Schlacher TA, Kloser RJ, Green MA, Barker growth rates of bamboo corals from the Gulf of Alaska. Geophys BA, Bax N, Brodie P, Schlacher-Hoenliger MA (2009) Impacts Res Lett 32:L04606. doi:10.1029/2004GL021919 of bottom trawling on deep-coral ecosystems of seamounts are Roberts JM, Wheeler AJ, Freiwald A, Cairns SD (2009) Cold-water long-lasting. Mar Ecol Prog Ser 397:279–294 corals. Cambridge University Press, Cambridge, UK Andrews AH, Cordes EE, Mahoney M, Munk K, Coale KH, Cailliet Sherwood OA, Edinger EN (2009) Ages and growth rates of some GM, Heifetz J (2002) Age, growth and radiometric age validation deep-sea gorgonian and antipatharian corals of Newfoundland of a deep-sea, habitat-forming gorgonian (Primnoa resedaefor- and Labrador. Can J Fish Aquat Sci 66:142–152 mis) from the Gulf of Alaska. Hydrobiologia 471:101–110 Sherwood OA, Scott DB, Risk MJ, Guilderson TP (2005) Radiocar- Brooke S, Young CM (2009) In situ measurement of survival and bon evidence for annual growth rings in the deep-sea octocoral growth of Lophelia pertusa in the northern Gulf of Mexico. Mar Primnoa resedaeformis. Mar Ecol Prog Ser 301:129–134 Ecol Prog Ser 397:153–161 Smith JE, Schwarcz HP, Risk MJ (2002) Patterns of isotopic Cairns SD (2007) Deep-water corals: an overview with special disequilibria in azooxanthellate coral skeletons. Hydrobiologia reference to diversity and distribution of deep-water scleractin- 471:111–115 ian corals. Bull Mar Sci 81:311–322 Sokal RR, Rohlf FJ (1981) Biometry, 2nd edn. Freeman, San Clementson LS, Parslow JS, Griffiths FB, Lyne VD, Mackey DJ, Harris Francisco GP, McKenzie DC, Bonham PI, Rathbone CA, Rintoul S (1998) Stuvier M, Pollach HA (1977) Discussion: reporting of 14C data. Controls on phytoplankton production in the Australasian sector of Radiocarbon 19:355–363 the subtropical convergence. Deep-Sea Res I 45:1627–1661 Thiagarajan N, Gerlach D, Roberts M, Burke A, McNichol A, Jenkins Davies AJ, Guinotte JM (2011) Global habitat suitability for B, Subhas A, Thresher RE, Adkins J (2013) Movement of deep- framework-forming cold-water corals. PLoS ONE 6:e18483 sea coral populations on climatic time scales. Paleoceanography Fallon SJ, Fifield LK, Chappell JM (2010) The next chapter in 28:227–236 radiocarbon dating at the Australian National University: status Thresher RE, Adkins J, Thiagarajan N (2011a) Modal analysis of the report on the single stage AMS. Nuclear Instr Methods Physicis deep-water solitary scleractinian Desmophyllum dianthus,on Res B 286:898–901 SW Pacific seamounts: inferred recruitment periodicity, growth Freiwald A, Roberts JM (eds) (2005) Cold-water corals and and mortality rates. Coral Reefs 30:1063–1070 ecosystems. Springer-Verlag, Berlin Thresher RE, Wilson NC, MacRae CM, Neil H (2010) Temperature Guinotte JM, Orr J, Cairns S, Freiwald A, Morgan L, George R (2006) effects on the calcite skeletal composition of deep-water Will human-induced changes in seawater chemistry alter the gorgonians (Isididae). Geochim Cosmochim Acta 74:4655–4670

123 38 Coral Reefs (2014) 33:31–38

Thresher RE, Tilbrook B, Fallon S, Wilson NC, Adkins J (2011b) Thresher RE, Williams A, Adkins A, Althaus F, Gowlett-Holmes K, Effects of chronic low carbonate saturation levels on the Alderslade P, Dowdney J, Fallon S, Trull T, Cho W, Gagnon A, distribution, growth and skeletal chemistry of deep-sea corals Staples D, McEnnulty F (in press). Depth effects on the and other seamount megabenthos. Mar Ecol Prog Ser 442:87–99 abundance, diversity and composition and biomass of megaben- Thresher RE, Rintoul SR, Koslow JA, Weidman C, Adkins J, Proctor thos on seamounts off Southern Tasmania. PLOS One C (2004) Oceanic evidence of climate change in southern Tracey DM, Neil H, Marriott P, Andrews AH, Calliet GM, Sanchez Australia over the last three centuries. Geophys Res Lett JA (2007) Age and growth of two genera of deep-sea bamboo 31:L07212 corals (family Isididae) in New Zealand waters. Bull Mar Sci Thresher RE, Adkins J, Fallon SJ, Gowlett-Holmes K, Althaus F, 81:393–408 Williams A (2011c) Extraordinarily high biomass benthic Yokoyama Y, Esat TM, Lambeck K, Fifield LK (2000) Last ice age community on Southern Ocean seamounts. Sci Rpts (Nature) millennial scale climate changes recorded in Huon Peninsula 1:119. doi:10.1038/srep00119 corals. Radiocarbon 42:383–401

123