Polar Biology

, Volume 35, Issue 7, pp 1027–1034

Combined effects of two ocean change stressors, warming and acidification, on fertilization and early development of the Antarctic echinoid Sterechinus neumayeri

  • J. A. Ericson
  • M. A. Ho
  • A. Miskelly
  • C. K. King
  • P. Virtue
  • B. Tilbrook
  • M. Byrne
Original Paper

DOI: 10.1007/s00300-011-1150-7

Cite this article as:
Ericson, J.A., Ho, M.A., Miskelly, A. et al. Polar Biol (2012) 35: 1027. doi:10.1007/s00300-011-1150-7


The effects of concurrent ocean warming and acidification on Antarctic marine benthos warrant investigation as little is known about potential synergies between these climate change stressors. We examined the interactive effects of warming and acidification on fertilization and embryonic development of the ecologically important sea urchin Sterechinus neumayeri reared from fertilization in elevated temperature (+1.5°C and 3°C) and decreased pH (−0.3 and −0.5 pH units) treatments. Fertilization using gametes from multiple males and females, to represent populations of spawners, was resilient to acidification at ambient temperature (0°C). At elevated temperatures, there was a negative interactive effect of temperature and pH on percentage of fertilization (11% reduction at 3°C). For cleavage stage embryos, there was a significant, but small reduction (6%) in the percentage of normal embryos at pH 7.5. For blastulae, a 10–11% decrease in normal development occurred in the +3°C treatments across all pH levels. Our results highlight the importance of considering the impacts of both temperature and pH in assessing the life history response of S. neumayeri in a changing polar ocean. While fertilization and development to the blastula stage were robust to levels of temperature and pH change predicted over coming decades, deleterious interactive effects were evident between these stressors at levels projected to occur by 2100 and beyond.


Ocean acidification Ocean warming Antarctica Echinoderm Early life history 


Sequestration of anthropogenic carbon dioxide (CO2) emissions into the ocean is driving ocean acidification, an increase in ocean surface pCO2(aq) and H(aq)+ ions, and a decrease in carbonate (CO3 (aq)2−) concentrations and seawater pH. Ocean surface pH has already decreased by approximately 0.1 pH units since the industrial revolution (Brewer 2009; Doney et al. 2009; Hoegh-Guldberg and Bruno 2010), and if emissions are not mediated, pH is predicted to decrease by 0.3–0.5 pH units by the year 2100 (Caldeira and Wickett 2003). Global ocean surface temperatures are also predicted to increase by an average 0.2°C per decade (IPCC 2007) due to rising atmospheric CO2 levels. The outcomes for marine biota in a changing ocean are of great concern and are also difficult to predict because the rate of change differs among regions.

The polar regions are likely to experience large-scale changes in ocean carbonate chemistry and seawater pH, with surface waters becoming undersaturated with aragonite in the twenty first century at the current rate of CO2 emissions (McNeil and Matear 2008). At the same time, the Antarctic Peninsula is one of the fastest warming regions on Earth (Hansen et al. 1999; Meredith and King 2005; Barnes et al. 2006; Clarke et al. 2007). Synergistic effects of increasing ocean surface pCO2 and temperature may be particularly important in polar oceans. Contemporary ocean change is already having deleterious impacts on Arctic and Antarctic organisms (Comeau et al. 2009; Hoegh-Guldberg and Bruno 2010; Lister et al. 2010). Thus, polar regions are suggested to be a bellwether for CO2-induced changes in other oceans (Fabry et al. 2009).

Antarctic organisms are stenothermal due to the stability of their environment over evolutionary time scales (Peck et al. 2004), and so polar marine invertebrates are particularly sensitive to perturbations in seawater temperature. The planktonic and benthic life stages of many Antarctic echinoderms and mollusks appear intolerant of even a small level of warming (Stanwell-Smith and Peck 1998; Peck 2005). Organisms such as the bivalves Laternula elliptica and Adamussium colbeckii shut down at temperatures 2–3°C above ambient (Peck 2005). These levels of warming are well within the range projected by 2100 (Intergovernmental Panel on Climate Change (IPCC) 2007).

Early life stages of most marine invertebrates are considered to be particularly vulnerable to ocean acidification because they have poor abilities to regulate internal pH and ion flux (Melzner et al. 2009). However, studies with a diverse suite of species indicate that the responses of the early life history stages to ocean acidification are hard to predict because responses are variable and may be species-specific, although some of this variation may be due to the disparate methods used (Byrne 2010, 2011a; Dupont et al. 2010). For polar species, single-stressor studies of the effects of ocean acidification on fertilization and development of the sea urchin Sterechinus neumayeri indicate that fertilization is robust to near-future-projected acidification (ca. pH 7.7) even at low sperm/egg ratios of 50:1 (Ericson et al. 2010). Development to the early pluteus stage in this species is also robust to reduced pH (ca. pH 7.7) (Clark et al. 2009; Ericson et al. 2010).

It is increasingly clear that the single-stressor perspective will not provide the insights needed to determine the responses of marine species to an environment simultaneously warming and increasing in pCO2. It is important to understand the synergistic effects of these factors on sensitive early life history stages which, due to the “developmental domino effect” (see Byrne 2011a), may be the weak link, a bottleneck for species’ persistence in a changing ocean. Here, we investigated the impacts of ocean warming and acidification on fertilization and early development in a keystone Antarctic echinoid, Sterechinus neumayeri. This species has a central trophic role in Antarctic nearshore marine ecosystems and is a model species for research due to its importance as a predator and grazer, and its abundant circumpolar distribution (Brey et al. 1995; Cowart et al. 2009). The response of embryos of S. neumayeri to environmental stressors has been investigated in several studies where negative effects of decreased salinity, increased UV-B, and exposure to metal contaminants have been demonstrated (King and Riddle 2001; Lamare et al. 2007; Cowart et al. 2009; Lister et al. 2010). Previous studies indicate that the optimum temperature range for development in S. neumayeri is 0.2–1.7°C, with significant increases in nonviable embryos at 3°C (ca. 5°C above ambient) (Stanwell-Smith and Peck 1998). The impacts of the two main ocean change stressors, warming and acidification on the early life history stages of S. neumayeri, are not known.

This study investigated the effects of increased seawater temperatures (+1.5–3°C) and decreased pH (−0.3 to 0.5 pH units) on fertilization and development through the first 3 days of development to hatching in Sterechinus neumayeri. Echinoderm development is well known to be highly sensitive to increased temperature with facilitation due to increased rates of development at low levels of warming, and developmental failure once thermal tolerance levels are exceeded (Staver and Strathmann 2002; Sheppard Brennand et al. 2010). We expected that warming would exert a more significant influence on fertilization and development in S. neumayeri than acidification, with failure of development in the highest temperature treatment. As temperature was expected to modulate the response to decreased pH, we also predicted that there would be negative synergistic effects of stressors, particularly at the highest temperatures tested.

Materials and methods

Organism collection and spawning

Adult Sterechinus neumayeri were collected from Ellis Narrows, near Davis Station, East Antarctica (68.616°S, 77.992°E) from 1 m depth during Dec–Jan 2010/11 within the peak summer spawning period for S.neumayeri (King and Riddle 2001). Sea urchins were held for 24 h in an aquarium at ambient pH (8.0) and temperature (−1 ± 1°C) and fed the macroalga Palmaria dicipens until used for experiments. In their shallow water habitat near Davis, S. neumayeri experience sea surface temperatures (SST) ranging from −1.7 to 2.0°C during the austral summer spawning season after the breakup of sea ice (Barnes et al. 2006; Data: 2007–2011: satellite source—GHRSST L4 Remote Sensing System MW IR OI SST; http://poet.jpl.nasa.gov/). We used these data to place our experiments in context for the region (see below).

Fertilization success

For each fertilization test, pooled gametes from multiple males and females (2–4 adults of each sex) were used to represent a population of spawners as might occur in nature. To induce gamete release, 0.5M KCl was injected into the coelom. Eggs were spawned into ambient seawater. Sperm was collected dry and kept in tubes on ice until needed. Gametes from each individual were examined for quality before use. Only gametes of high quality, i.e., sperm active and eggs with a uniform profile, were used. Eggs were gently mixed and divided into 700 ml vials at concentrations of ca. 40 eggs/ml as determined from counts of eggs in six 1 ml aliquots of the egg suspension. The eggs were left to acclimate in experimental filtered seawater (FSW, see below) for 15 min before the addition of sperm. The optimal sperm-to-egg ratio (800:1) was determined from a previous study (Ericson et al. 2010). To achieve this ratio, the number of sperm in the semen sample was counted using a hemocytometer and 1 μl of the semen was placed in 10 ml of experimental water and briefly activated. The appropriate amount of this diluted sperm solution to achieve the target sperm/egg was added to the egg suspension. This approach was taken to avoid problems of variable sperm aging as in the dilution series approach taken in standard protocols for sea urchin fertilization toxicity tests (US EPA 2002). After 4 h, ten samples were removed from each treatment vial (temperature: 0°C, +1.5°C, +3°C, pHT: 8.0, 7.7, 7.5) (Table 1) to score percent fertilization based on enumeration of 40–50 eggs from each sample. Fertilization was determined by the presence or absence of a fertilization envelope.
Table 1

Mean experimental seawater pHT conditions for source water measured 12 times through the study determined using CO2SYS (Pierrot et al. 2006; “Materials and methods”)





pH total


pH total


pH total


pH 8.0 (control)







pH 7.7







pH 7.5







The pHNIST measured on the day is provided for comparative purposes (n = 12, all standard errors were <0.001)

Embryonic development

Fertilization was achieved as above. Embryos derived from multiple parents were transferred into 250 ml vials at concentrations of 5 embryos/ml and reared to the 20 h cleavage (9 replicates per treatment) and the 3 day blastula (8 replicates per treatment) stages. At 20 h post-fertilization, the percentage of normal cleavage stage embryos was scored based on enumeration of 40–50 randomly sampled embryos per vial. At 3 days post-fertilization, the percentage of normal and abnormal hatched blastulae was scored in 40–50 embryos in each treatment. The vials were discarded after scoring. Embryos that had irregular profiles, irregular cell division, or arrested development (fertilization envelope, no cleavage) were considered delayed or abnormal, as illustrated in Ericson et al. (2010).

Adjustment of seawater pH and temperature and measurement of seawater parameters

Freshly filtered (1 μm) ambient seawater (FSW) was used for experiments, and FSW parameters were measured 12 times during experiments (mean: pHT = 7.99 ± 0.01; salinity = 34.21 PSU ± 0.03; total alkalinity = 2,321 ± 4.4 μmol/kg; DIC = 2,219 ± 5.7 μmol/kg). The aquarium system maintained three pH/pCO2 levels (pH 8.0/450 ppm, pH 7.7/850 ppm, pH 7.5/1,370 ppm), within the range projected for Antarctic waters over coming decades (IPCC 2007; McNeil and Matear 2008). The highest pCO2 level (1,370 ppm) albeit higher than near-future projections (e.g., Cao and Caldeira 2008) was used to determine a potential threshold level of this stressor. Experimental pH/pCO2 was achieved using HORIBA STEC SEC-E50 CO2 mass flow controllers and HORIBA STEC PE-D20 power supply control units. Before the gas mixture was injected into three 12-chambered perspex equilibrators containing recirculating FSW, a gas sample was piped through a Telaire 7001 CO2/Temperature monitor to ensure that it was at the desired CO2 ppm level. The gas mixture was then injected directly into each chamber of the equilibrators as FSW moved and allowed to equilibrate for ~24 h. Once the pH was stable at the desired level, it was used for experiments. Seawater for each treatment level was placed in sealed vials with no head space to minimize loss of CO2. The vials were kept at experimental temperatures of 0°C, 1.5°C, and 3°C using temperature-controlled cabinets (TCCs) monitored by data loggers (i-buttons, Thermodata). The 3°C treatment is 1–2°C above the maximum summer SST that the sea urchins would experience during their spawning season at the collection site.

Seawater samples (250 ml) were fixed with 100 μl of a saturated mercuric chloride solution and used to determine carbonate chemistry parameters. Total alkalinity (TA) was measured using an open-cell potentiometric titration, and dissolved inorganic carbon (DIC) was measured using the coulometric method after Dickson et al. (2007). Certified reference material from the Scripps Institution of Oceanography was used as a reference and indicates a measurement accuracy and precision of ± 2 μmol/kg for both parameters. The TA and DIC were used to calculate the pH of the water on the total scale (Table 1), using the equilibrium constants of Merhbach, as modified by Dickson and Millero (Dickson et al. 2007). The pH of the experiments was monitored in real time using a temperature-corrected pH meter (WTW 3400i) (see Table 1). The pH probes were calibrated frequently using NIST buffers of 4.0 and 7.0. To facilitate comparison with previous studies that report pH on the NIST scale, data for both pH measured on the day (pHNIST) and pHT calculated on the total scale using TA and DIC in CO2SYS (Pierrot et al. 2006) for the same seawater are presented in Table 1. In the text below, pH levels are indicated on the total scale. As expected, the pHT determinations were lower than the pHNIST readings (Zeebe and Wolf-Gladrow 2005). Random measurement of pH in the containers at 4 and 20 h indicated no change in pH, whereas a slight change was observed at 70 h in containers with hatched blastulae (Table 2).
Table 2

pHT and temperature conditions in experiments with Sterechinus neumayeri at the outset (fertilization) and after 70 h at the blastula stages (n =4)

Treatment start conditions

Blastula stage (70 h)

pH (±SE)

Temp °C (±SE)

pH 7.99, 0.0°C

8.03 (±0.03)

−0.45 (±0.29)

pH 7.99, 1.5°C

7.98 (±0.003)

0.9 (±0.29)

pH 8.0, 3.0°C

8.13 (±0.05)

2.9 (±0.05)

pH 7.73, 0.0°C

7.81 (±0.05)

−0.5 (±0.21)

pH 7.74, 1.5°C

7.70 (±0.02)

1.6 (±0.02)

pH 7.74, 3.0°C

7.76 (±0.04)

3.0 (±0.15)

pH 7.53, 0.0°C

7.53 (±0.03)

−0.3 (±0.24)

pH 7.54, 1.5°C

7.60 (±0.02)

1.8 (±0.09)

pH 7.54, 3.0°C

7.63 (±0.03)

3.4 (±0.14)

Statistical analyses

Percentage data on fertilization success and embryo development were arcsine square root transformed prior to analysis by two-way analysis of variance (ANOVA) with temperature and pH as fixed orthogonal factors. All assumptions of ANOVA were met. Tukey’s post hoc tests were carried out to examine differences among pH treatments and temperature levels. Statistical analyses were carried out using JMP (SAS Inc).


Fertilization success

High temperature and low pH had a significant negative effect on the percentage of fertilization (ANOVA: Temp: F2,15 = 4.13, P < 0.05; pH: F2,15 = 21.04, P < 0.001) (Fig. 1; Table 3), and there was a significant interaction between the two stressors (ANOVA Temp*pH: F4,15 = 4.13, P < 0.005). Negative effects were evident at elevated temperature with an 11% reduction in fertilization in the 1.5°C/pH 7.5 and 3°C/pH 7.5 treatments (Fig. 1). Tukey’s post hoc tests indicated that the elevated temperature and low pH treatments differed significantly from all other stressor combinations. At ambient temperature, there was no effect of pH across all levels. Elevated temperatures were not deleterious to fertilization at pH 7.7–8.0.
Fig. 1

Percentage of fertilization in Sterechinus neumayeri in three temperature and three pH levels in all combinations. Elevated temperature and decreased pH (pH 7.5) had a negative effect on fertilization at the higher temperatures and with a significant interaction between stressors (n = 10)

Table 3

ANOVA of percent data for fertilization, normal cleavage stage embryos, and normal blastulae reared in different temperature/pCO2 combinations




Sum of squares









(0, 1.5) > 3






(8.0, 7.7) > 7.5

Temp × pH*






TK: Temp × pH 0/8.0, 0/7.7, 0/7.5, 1.5/8.0, 1.5/7.7, 3/8.0, 3/7.7    3/7.7, 1.5/7.5, 3/7.5

Embryo cleavage












(8.0, 7.7) (7.7, 7.5) 8.0 > 7.5

Temp × pH












(0, 1.5) > 3







Temp × pH






* Designates significance. For the significant interaction, underlined treatments do not differ

Embryonic development

Decreased pH had a negative effect on early embryos (ANOVA pH: F2,15 = 4.22, P < 0.02), while there was no effect of elevated temperature (Table 3). Tukey’s post hoc tests indicated that there was a significant, but small reduction (6%) in the percentage of normal cleavage stage embryos at pH 7.5 (Table 3; Fig. 2). Conversely, elevated temperatures had a negative effect on development to the blastula stage (ANOVA Temp: F2,15 = 41.6, P < 0.001), while there was no effect of decreased pH (Table 3). At +3°C, there was a 10–11% decrease in normal embryos across all pH levels (Table 3; Fig. 3).
Fig. 2

Percentage of normal cleavage stage embryos of Sterechinus neumayeri in three temperature and three pH levels in all combinations. Decreased pH (pH 7.5) had a negative effect on cleavage with a 6% reduction compared with controls (n = 9)

Fig. 3

Percentage of normal blastulae of Sterechinus neumayeri in three temperature and three pH levels in all combinations. Elevated temperature had a negative effect on development (n = 8)


There was no effect of pH across all levels at ambient temperature, as found in a previous study (Ericson et al. 2010), but when temperature was added to the stressor mix, deleterious effects of both acidification and warming were evident. This highlights the importance of considering multiple stressors in assessing the life history response of Sterechinus neumayeri in a changing polar ocean.

Elevation of seawater temperatures was not deleterious to fertilization in S. neumayeri at pH 7.7–8.0, perhaps due to enhanced sperm metabolism. Other studies have suggested that sperm swimming speed increases in warmer water due to a decrease in the viscosity of seawater, which enhances the likelihood of collision with eggs (Mita et al.1984; Kupriyanova and Havenhand 2005). The negative effect of the pH 7.5/3°C treatment may be due to this treatment exceeding the metabolic/energetic capabilities of the sperm of S. neumayeri.

The impacts of ocean warming and acidification on fertilization in marine invertebrates are complex, and vary between species, although much of this variation is likely attributed to differences in experimental methods, particularly with the use of single dam-sire crosses vs. the spawner population approach (Byrne 2011a). Previous studies have shown that fertilization in echinoderms is generally robust to increases in seawater temperature of 4–6°C above ambient (Reviews, Byrne 2010, 2011b), while negative effects of pH on fertilization are usually only seen below pH 7.4 (Rupp 1973; Bay et al. 1993; Kurihara and Shirayama 2004; Carr et al. 2006; Ericson et al. 2010). Elevated seawater pCO2 has a narcotic effect on sperm decreasing respiration and flagellar motility (Christen et al. 1982). Sperm performance is affected by the narcotic effect of increased pCO2, the stimulatory effects of increased temperature, sperm activating peptides in egg jelly, and chemicals released by eggs as well as variable dam-sire compatibility due to gamete recognition protein polymorphisms (Palumbi 1999; Darszon et al. 2008; Krug et al. 2009). These factors do not act in isolation, and their interactive effects provide a significant challenge to our understanding of the mechanisms underlying the response of echinoid fertilization to ocean change stressors.

Decreased pH had a negative effect on early embryos, while there was no effect of elevated temperature. Conversely, elevated temperatures had a negative effect on development during the blastula stage, while there was no effect of decreased pH with a 10–11% decrease in normal embryos across all pH levels at +3°C. Similarly, Stanwell-Smith and Peck (1998) found that elevated temperature, albiet at more extreme levels (+5°C), resulted in a 13% decrease in viable S. neumayeri embryos. In previous experiments on S. neumayeri conducted at ambient temperature, development up to the gastrula stage was compromised, but only at extreme pH levels (<pH 7.3) (Ericson et al. 2010).

The response of marine invertebrates to temperature and pH stressors often differs between life stages (Wright et al. 1983; Byrne et al. 2009, 2010; Byrne 2011b). In other echinoderms, development is noted to be robust to temperatures ca. 4°C above ambient (Rupp 1973; Fujisawa 1989; Sewell and Young 1999; Benitez Villalobos et al. 2006; Byrne et al. 2009, 2010; Sheppard Brennand et al. 2010). Sterechinus neumayeri development was robust to decreases in seawater pH 0.5 units below ambient, similar to that determined for other sea urchins (Carr et al. 2006; Byrne et al. 2009, 2010, 2011; Catarino et al. 2011). A recent study on the subantarctic sea urchin Arbacia dufresnei found that there was no increase in developmental abnormalities in larvae reared at pH levels down to pH 7.4 (Catarino and Ridder 2011).

In the present study, ca. 87% of S. neumayeri embryos developed normally to the blastula stage across all pH and temperature treatments. The relatively robust response of most embryos to warming and acidification may be due to the presence of maternal protective factors such as heat shock stress proteins that are loaded into sea urchin eggs during oogenesis (Hamdoun and Epel 2007). These factors protect early embryos prior to switching on of the zygotic genome around blastulation.

It is not known how a decrease in fertilization (11%) and a reduction in normal embryos and blastulae (6–10%) would translate in terms of survival and recruitment during later life stages. Mortality is high during the planktonic life stages of marine invertebrates (Rumrill 1990; Lamare and Barker 1999). Thus, even the apparently small negative effects seen here on fertilization and development in S. neumayeri may have a significant flow on effect on population size in the long term in a changing ocean. Moreover, the embryological stages of polar species may be particularly at risk from climate-driven ocean change due to their slow developmental rates and long residence times in the water column (Pearse et al. 1991). The ability for echinoderms to adapt to increasing ocean pCO2 and temperature over 50–100 year time scales is unknown; however, phenotypic and genotypic variation in the response of echinoid larvae exposed to near-future pCO2 scenarios has been demonstrated in progeny derived from separate dam-sire crosses, suggesting the potential for acclimatization and adaptation (Sunday et al. 2011).

To date, single-stressor studies on the impacts of ocean acidification on early development of polar and subpolar echinoderms (e.g., Clark et al. 2009; Ericson et al. 2010; Catarino and Ridder 2011) indicate that high latitude species may not be more vulnerable to ocean acidification than temperate and tropical species. We observed deleterious interactive effects of warming and acidification on S.neumayeri, and it is clear that more comparative studies involving both of these ocean change stressors across regions are needed to understand vulnerabilities.

Environmental considerations

The response of marine invertebrate developmental stages to perturbations often reflects the environmental history of the maternal parent (Hamdoun and Epel 2007). The resilience of the early developmental stages of S. neumayeri to increased temperatures and pCO2 in this study may reflect the environmental history (phenotypic plasticity) of adults from the shallow coastal waters around Davis Station. In this habitat, the gametes, embryos, and larvae of this species experience seasonal fluctuations in SST and temperatures up to ca. 3°C, albeit for short seasonal periods. These temperatures are warmer than those experienced in the deeper water habitat where this species occurs elsewhere in Antarctica (e.g., McMurdo Sound; −1.5 to −2.0°C, 20 m) (McClintock 1994; Barnes et al. 2006; Lamare et al. 2007; Ericson et al. 2010). As seawater pH is strongly influenced by temperature and seasonal biological drawdown of CO2 (Gibson and Trull 1999), local S. neumayeri and its progeny are also likely to be exposed to large seasonal fluctuations in seawater pH. To make regional comparisons on the response of the gametes and embryos of broadly distributed species like S. neumayeri to changing ocean conditions, experiments need to be placed in context with the environmental history of the parents and current baseline conditions (Sorte et al. 2011). For the population of S. neumayeri in the shallow waters near Davis, if the baseline of surface sea conditions change in line with general projections (i.e., +2°C, −0.4 pH units, by 2100), then the developmental stages of this species may experience conditions that exceed their tolerance limits.


This work was supported by a grant from the Australian Antarctic Division (AAD). Thanks to AAD staff and students who provided assistance and logistical support including Rob King, Steve Whiteside, John van den Hoff, Andrew Bryant, Clive Strauss, Charmaine Alford, Kathryn Brown, Lara Marcus, Sarah Payne, Bianca Sfiligoj, Claire Wallis, Debbie Lang, Jane Wasley, Dougie Gray, and Leigh Hornsby.

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • J. A. Ericson
    • 1
    • 2
  • M. A. Ho
    • 1
  • A. Miskelly
    • 1
  • C. K. King
    • 3
  • P. Virtue
    • 4
  • B. Tilbrook
    • 5
    • 6
  • M. Byrne
    • 1
    • 7
  1. 1.School of Medical SciencesUniversity of SydneySydneyAustralia
  2. 2.University of OtagoDunedinNew Zealand
  3. 3.Australian Antarctic DivisionKingstonAustralia
  4. 4.Institute for Marine and Antarctic StudiesUniversity of TasmaniaHobartAustralia
  5. 5.Centre for Australian Weather and Climate Research, a partnership between CSIRO and the Bureau of MeteorologyHobartAustralia
  6. 6.Antarctic Climate and Ecosystems Cooperative Research CentreHobartAustralia
  7. 7.School of Biological SciencesUniversity of SydneySydneyAustralia

Personalised recommendations