Coral Reefs (2008) 27:227–236 DOI 10.1007/s00338-007-0316-9
REPORT
High rates of nitrogen fixation (acetylene reduction) on coral skeletons following bleaching mortality M. Davey Æ G. Holmes Æ R. Johnstone
Received: 8 March 2007 / Accepted: 12 September 2007 / Published online: 16 October 2007 Springer-Verlag 2007
Abstract Nubbins of the coral Acropora aspera were artificially bleached and nitrogen fixation (acetylene reduction) rates were measured on the developing epilithic communities. Seasonal comparisons were made between corals that died in summer of heat stress and corals that died in winter from natural cold stress. Rates of acetylene reduction from artificially bleached corals peaked at 26.66 nmol cm–2 h–1 2 weeks after summer mortality, while rates from natural winter mortality peaked at 18.07 nmol cm–2 h–1 12 days after coral death. Comparative rates of acetylene reduction taken from live corals and coral rubble ranged between 0.56 and 1.16 nmol cm–2 h–1, and 0.15 and 12.77 nmol cm–2 h–1, respectively. N2-fixation rates from dead corals were up to 30 times greater than those measured on live corals. The observed increase in N2-fixation from dead corals may increase the availability of nitrogen for use in trophic processes within the reef for an extended period following the initial mortality event. If the spatial scale over which coral mortality has occurred in past thermal bleaching events is considered the ramifications of such an increase may be substantial. Keywords Nitrogen Carbon Bacteria Algae Trophic interactions
Communicated by Biology Editor M.P. Lesser. M. Davey (&) G. Holmes R. Johnstone Centre for Marine Studies, University of Queensland, St Lucia, QLD 4072, Australia e-mail:
[email protected]
Introduction It is now widely accepted that coral reefs rely on considerable internal recycling and nutrient conservation in order to meet their needs in oligotrophic oceanic waters (Ryther 1969; Wilkinson et al. 1984; Alongi et al. 2006). Both Johannes et al. (1972) and Webb et al (1975) observed that the export of nitrogen (N) from reef waters can far exceed the input from surrounding oceans and suggested that internal nitrogen transformations on various reef substrates were responsible. Of these, Nitrogen (N2) fixation was postulated as a critical process in supplying dissolved inorganic nitrogen (DIN; i.e., the combined forms used in plant growth) and thus balancing the observed loss. Subsequent investigations have determined that N2-fixation may be an important contributor to the DIN pool and to the overall N budget of coral reefs (e.g., Crossland and Barnes 1976; Wilkinson et al 1984; Capone et al. 1992). The process of N2-fixation is carried out by an array of micro-organisms and has been measured from many reef surfaces including carbonate sediments (Wilkinson et al. 1984; Capone et al. 1992), algal-dominated reef crests (Larkum et al. 1988), living corals (Crossland and Barnes 1976; Shashar et al. 1994a, b; Lesser et al. 2004), coral rubble (Wilkinson et al. 1984; Larkum et al. 1988), and new coral skeletons (Larkum 1988). Notably however, the fate of DIN produced via N2-fixation from each of these substrates can vary considerably. For example, N fixed by coral associated micro-organisms may be used by corals for growth and maintenance (Ritchie and Smith 2004), trapped by coral mucus and subsequent transported to reef sediments (Wild et al. 2004), and/or lost from the reef via export. Similarly, N2-fixation which occurs readily in many reef sediments may contribute significantly to the overall N budget (Johnstone et al. 1989). However, the amount of N
123
228
delivered to the water column and areas remote to the sediments remains uncertain but may be limited in terms of direct DIN production and release (Johnstone et al. 1990; Charpy-Roubaud et al. 1996). Death of corals as a result of thermal bleaching may enhance N production, through increased development of microbial epiphytic and epilithic communities on coral skeletons. If N2-fixation is facilitated within these new communities, the amount of new N delivered to the reef is unclear but may be significant. Furthermore, the poor conditions that may prevail following widespread death of corals have the potential to disrupt the usually high level of nutrient conservation that occurs on coral reefs. This may further enhance the availability of nutrients for use in other trophic processes. Despite the prevalence of N2-fixation, this and other ‘new’ inputs of N (e.g., upwelling) may not be sufficient to meet biological demand, and as a result N availability can limit plant production (Williams and Carpenter 1988; Furnas et al. 2005). Hence any sudden alteration of supply stands to potentially alter both productivity and trophodynamics within the ecosystem involved. This supply-response model has been observed in a wide range of ecosystems including temperate, cold-temperate (e.g., Larsson and Hagstrom 1982; Bianchi et al 2000), and tropical (e.g., Banner 1974; Hunter and Evans 1995). By corollary, the nutrient processes on reefs and the microbial communities that underpin them are vulnerable to disturbance. Past perturbations have altered microbial communities to such a degree that dramatic changes in nutrient remineralisation and trophodynamic processes (in particular N related processes, Paerl 1998) have occurred. Accordingly, any disturbance that alters the structure of microbial communities on reef surfaces has the potential to dramatically alter the N processes that can occur on those substrates, including corals (e.g., Mitchell and Cuet 1975). Coral reefs currently face extensive mortality that is likely to continue into the foreseeable future (Houghton et al. 2001) and as a consequence it could be expected that the extent of new surfaces from dead corals will increase correspondingly. In view of recent bleaching events this may represent a large percentage of living coral encompassing up to 100% of a given reef (Wilkinson 2000). Such death of coral over large spatial scales will increase the surface area available for microbial colonisation and potentially facilitate microbial transformations of N (Paerl 1998). This may ultimately result in a greater availability of combined nitrogen compounds in the immediate vicinity of the coral surface, and potentially beyond. For example, locally, greater N availability may stimulate epiphytic and epilithic algae growth on the coral surfaces, which in turn may further enhance nitrogen fixation through the creation of niche environments (Paerl and Prufert 1987). If
123
Coral Reefs (2008) 27:227–236
concentrations of N in the internal waters of reefs are increased this may stimulate phytoplankton productivity within the water column (Furnas et al. 2005), which in turn may increase the deposition of carbon and nitrogen to reef sediments and alter processes occurring therein (Alongi et al. 2006). Excess nutrients may additionally be exported to surrounding oceanic waters where they may stimulate phytoplankton productivity (Webb et al. 1975). The magnitude of the delivery, if any, and the efficiency of recycling processes within the reef post-mortality will ultimately determine changes in growth of opportunistic phytoplankton and algae species. The effects of nutrient inputs on coral reefs are confounded however, by a myriad of factors that can influence the response of the ecosystem. As a result this has been the subject of much debate in the scientific community and we are yet to understand all the interlinking factors that may result in, for example, changed trophic interactions following nutrient inputs. Short term nutrient pulses to coral reefs can promote a significant increase in macroalgal photosynthesis and growth (Schaffelke and Klump 1998). Nutrient inputs to reefs over longer periods have resulted in substantial changes in the trophic complexities of these reefs (Hunter and Evans 1995). Conversely the ENCORE experiment demonstrated that in some reef systems nutrient additions may have no significant effect on macroalgal growth but they can have a detrimental effect on coral calcification and growth (Koop et al. 2001). Regardless and as noted, if we take into account the scale of mortality in past bleaching events (Wilkinson 2000) and the likely increased frequency of severe bleaching events in the future (Houghton et al. 2001), changed N dynamics on coral surfaces resulting from changes in microbial community structure may alter standing stocks of N. The increase of algal biomass and processes such as atmospheric nitrogen fixation may ultimately change trophic relationships and accelerate reef degradation. To date, reports on increases in nutrient concentrations on reefs which have contributed to changes in trophic structure and interactions, have only considered the effects allochthonous inputs (e.g., sewage, Banner 1974; eutrophication, Costa et al. 2000). In addition, the cumulative affect of episodic nutrient inputs, such as those that occur as a result of flood discharges, have also been implicated in reef degradation (e.g., Brodie and Furnas 1996). Notably, coral death itself may lead to an increase in autochthonous nutrient inputs at concentrations sufficient to cause a decline in reef health. This study examined the potential for changes in the processes undertaken on surfaces produced by coral death and how these may influence the nutrient dynamics
Coral Reefs (2008) 27:227–236
on individual corals, and considered the potential ramifications these changes may have for the wider coral reef ecosystem. With specific focus on N2-fixation, the study assessed the extent to which this process was altered on dead coral surfaces relative to living corals. A critical consideration in assessing the experimental results is the large spatial scales over which coral mortality has occurred in past bleaching events (Wilkinson 2000, 2002, 2004). Significant changes in process rates and their spatial distribution relative to non-bleached or killed reefs may potentially provide enough new nitrogen to substantially alter the overall ecosystems trophodynamics and integrity. In order to effectively manage reefs subject to significant coral mortality it is important that we develop an understanding of the function of these postmortem reefs so that we can better adjust our inputs and impacts accordingly.
Materials and methods Study area Field experiments were carried out on the platform reef surrounding Heron Island (ca. 800 · 200 m), which is part of the Capricorn-Bunker group located 70 km offshore on the continental shelf. The reef lies on the Tropic of Capricorn at the southern end of the Great Barrier Reef (23260 S, 151550 E). In situ experiments were carried out approximately 150 m (from the spring high tide mark) off the south-west facing beach. Water movement at Heron Island reef is largely due to wave generated currents (Gourlay and Colleter 2005); however tides, wind and the specific morphology of the reef also have an effect.
Collection of experimental organisms Ten fragments of the experimental organisms Acropora aspera were collected from each of four separate colonies from the reef flat on Heron Island. Corals were broken into nubbins (2–3 cm long pieces) and cemented (Knead-it Aqua1, Selleys Australia) on to plastic racks. The racks were placed into experimental tanks fed with water direct from the reef surrounding Heron Island. Three of the six tanks were fitted with custom designed thermostatically controlled heaters to raise temperatures in order to simulate thermal bleaching. Corals were acclimated for 1 week prior to inducing bleaching. All tanks were covered with shade cloth to reduce irradiance to those common to the collection area. Nubbins were divided randomly between the control and heated tanks.
229
Induced mortality of corals via heat stress Mortality of corals was induced via thermal bleaching. In the three experimental tanks fitted with custom designed thermostatically controlled heaters the temperature was raised to 32–33C over a period of 5 days following Wilkinson et al. (1999). Since it was intended to induce mortality of corals, the upper temperatures were used. The temperature increase employed was still in accordance with literature values of past bleaching events resulting in coral mortality.
Mortality of corals via cold stress Winter experiments were carried out in July and August of 2003. In August 2003, unusually cold wintry weather resulted in mortality of reef flat corals on Heron Island (Hoegh-Guldberg et al. 2005). Acropora species were particularly vulnerable as these are the most abundant species on Heron Island (Mather and Bennett 1984). In this instance, ten live and ten recently dead pieces were collected from each of four separate colonies of A. aspera.
Determination of acetylene reduction rates Rates of nitrogen fixation were estimated using the acetylene reduction assay (ARA) (Stewart et al. 1967). Incubations were performed in gas-tight PerspexTM chambers, which were sealed at the top and bottom with non-reactive plastic bungs and o-rings. The top bung contained a gas-tight rubber septum to allow needle penetration for acetylene introduction and subsequent sampling. On each experimental day coral nubbins were collected from experimental racks, placed into incubation chambers (6), covered with filtered (0.22 lm, Whatman GF) seawater (160 ml), with air (40 ml) in the head space. Twenty percent of the air within the chamber was replaced with acetylene as per Larkum et al. (1988) and the chambers agitated by hand for several minutes. Experimental chambers were then placed in an aquarium with a custom designed, waterproof, battery operated magnetic stirrer, which in turn made the stirring mechanism within each chamber rotate gently. The temperature in the aquarium was maintained at 22C for winter experiments and 28C for summer experiments for the duration of the incubations. These temperatures are within the averages observed on the reef flat at Heron Island (Dove 2004; Hoegh-Guldberg et al. 2005). Irradiance in the aquaria was maintained at [1,000 lmol cm–2 s–1 for the duration of incubations (6 h), using an Aqualina 120 double fluorescent light source (Aqualina, USA).
123
230
Coral Reefs (2008) 27:227–236
Gas samples (5 ml) were taken from each replicate experimental and live coral sample, at 0, 2, 4, and 6 h following introduction of acetylene gas on each of the sampling days. Gas samples were collected in VacutainersTM (Becton Dickinson, USA) and the volume of gas drawn out was replaced with the same volume of ambient air. All samples were immediately refrigerated at 4C until analysis. Ethylene concentrations were determined using a Shimadzu Gas Chromatograph GC-17A (Shimadzu, Tokyo, Japan) equipped with a flame ionization detector (FID) and a 1.8 m ·3 mm Porapak N (stainless steel) column maintained at 90C. Ultrahigh purity (UHP) helium (BOC gases, Australia) was used as the carrier gas. Instrument sensitivity and linearity were determined using a calibrated set of UHP ethylene standards (BOC Gases, Australia). The flame ionisation detector was maintained at 80C with hydrogen flow at 35 ml min–1. The support gas was instrument grade air (BOC Gases, Australia) kept at a flow rate of 400 ml min–1. Three replicates (100 ll) were injected for each sample using a gas syringe (Hamilton, USA). Acetylene reduction rates were estimated using a modification of the calculations outlined in Larkum et al. (1988): PA Sample=PA Standard ½Std GPV SC ¼ acetylene reduction nmol where PA sample is the peak area response for ethylene (C2H4) for a standard volume of sample, PA Standard is the linear equation of the known standards peak area responses for the same volume (in this case standards run were 2,000, 1,000, 200, 20 and 2 ppm), [Std] is the concentration of the ethylene standard used expressed in nanomolar per milliliter, GPV is the gas phase volume (the total volume of the gas head space of the assay vessel in milliliter), and SC is the solubility correction for ethylene in aqueous phase. Where SC is M/X is 1 + (a · A/B) and, a is the Bunsen Coefficient for ethylene at the appropriate temperature and salinity (Breitbarth et al. 2004), A is the volume of the aqueous phase of the assay vessel, B is the volume of the gas phase (or GPV) of the assay vessel. The linear trend of the time series was then determined to give acetylene reduced in nanomolar per minute, which was then normalized for surface area to give acetylene reduced in nanomolar per minute per square centimeter. Since a mass spectrometer could not be accessed for this study it was not possible to conduct a parallel 15N rate determination for calibration of the acetylene inhibition results. In order to place the measured rates in context and for comparisons the results are quoted as acetylene reduction rates. Also, in the absence of a specific conversion factor for dinitrogen
123
fixation from the acetylene reduction measurements, the literature values 3 (Hardy et al. 1968) and 3.45 (Larkum et al. 1988) were used. Whilst it is acknowledged here that the specific conversion factor is dependent upon the composition of the associated bacterial populations, the characteristics of the substrate investigated and the prevailing environmental conditions (Postgate 1982; Larkum 1988; Larkum et al. 1988), the range of reported conversion factors does provide a basis for assessing the potential significance and range of N2-fixation rates that may occur at a given location (Larkum 1988). Determination of C:N ratio of epilithic surface community Live corals and dead corals collected on Days 0, 4, 8, 12, 14 and 21 after coral mortality, were dried at 65C for [48 h and crushed using a mortar and pestle. Samples for C analysis were decalcified using 2 M HCl (Aristar grade, 35%), rinsed with ultrapure water (Milli-Q) and dried at 65C for [48 h. A LECO CNS-2000 analyser (LECO Corporation, MI, USA) was used to determine total N and C content from ground samples, and organic C content form ground acid treated samples. It is noted that according to Yamamuro and Kayanne (1995) values for both C and N obtained from samples rinsed in ultrapure water may be up to 50% lower than those obtained by other methods.
Determination of surface population d15N natural abundances Changes in d15N natural abundance were used to verify acetylene reduction rates and indicate the fate of any fixed nitrogen. Samples for d15N analysis were not acid treated as acidification is known to affect d15N values (Bunn et al. 1995). Small volumes of the samples were oxidized, and the resultant N2 was analysed for d15N signatures using a continuous flow-isotope ratio mass spectrometer. d15N results are reported relative to the atmospheric standards. Values were calculated according to: d15N% = (Rsample/ Rstd) – 1 · 1,000, where R = 15N/14N (Clapcott and Bunn 2003). All samples were analysed at the Centre for Riverine Landscapes, Griffith University, Australia.
Statistical analysis Differences between experimental and control samples were tested for significance using a Repeated Measures ANOVA with 95% confidence interval (Statistica 6—StatSoft Inc., Oklahoma, USA). Where data failed a Cochran C, Hartley,
Coral Reefs (2008) 27:227–236
231
Bartlett test of homogeneity of variance, data was transformed using log + 1 to achieve normality. Scheffe´ post-hoc tests were performed to analyse specific significances within the results.
Results Significantly higher acetylene reduction rates were observed between control corals (live and coral rubble) and dead coral (days 0 to 21 and 11 months) for both seasons (Fig. 1). Summer acetylene reduction rates ranged from 1.08 to 26.66 nmol cm–2 h–1 for surface communities (biofilm) on dead coral, compared to 0.48–0.98 nmol cm–2 h–1 for live corals (Fig. 1a). All rates from biofilms exceeded those observed from corresponding live corals. Significant differences were observed at days 8, 12 and 21, and at 11 months after coral mortality (F = 42.93,
Acetylene Reduction (nmol cm-2 hr-1)
30
(a)
25 20 15 10 5 0
Acetylene Reduction (nmol cm-2 hr-1)
0
30
4
8
12
21
11 M
Rubble
(b)
25
p \ 0.001). The highest acetylene reduction rate measured from dead coral surfaces occurred at Day 12 after induced mortality. In winter experiments, acetylene reduction rates ranged from 1.48 to 19.07 nmol cm–2 h–1 for biofilms on dead corals, compared to 0.30–1.16 nmol cm–2 h–1 for live corals (Fig. 1b). Significant differences between biofilms on dead coral and live corals were observed on day 12 (F = 6.23, p \ 0.05) after mortality. Comparisons of summer and winter rates showed no significant difference in rates from live corals. There were significant differences between biofilms on dead corals (F = 5.69754, p \ 0.05), however post-hoc analysis revealed these were only at days 8 and 21. Interestingly, rates were not significantly different on day 12 following mortality. Rates of nitrogen fixation inferred from acetylene reduction conversion factors from the literature resulted in a potential range of summer nitrogen fixation rates of 0.71 to 13.79 mg N2 m–2 day–1 (Table 1). The potential range of winter nitrogen fixation rates was 0.79–7.97 mg N2 m–2 day–1 (Table 1). d15N natural abundances plotted against acetylene reduction rates are shown in Fig. 2. The decrease in d15N immediately preceding an increase in acetylene reduction verifies the rate measurements achieved by ARA, because decreasing d15N abundances are indicative of atmospheric nitrogen fixation activity. As such, the decreasing d15N abundances may also indicate an increase in nitrogen fixing entities within biofilms on dead coral over time (Gu and Alexander 1993). In both instances this is due to fractionation of nitrogen that occurs during nitrogen fixation. Furthermore, the decrease in acetylene reduction witnessed immediately following the highest rates (day 12) in both summer and winter confers with the lower d15N natural abundances on day 21, which implies that the availability of N2 into the community may be limiting.
20 15
Table 1 Acropora aspera. Estimated nitrogen fixation (mg N2 m–2 day–1) calculated from acetylene reduction rates (in parenthesis; nmol cm–2 h–1) as a daily rate averaged across the experimental period for dead coral biofilms using literature conversion rates
10 5
Treatment 0 0
4
8
12
21
3.45a
Rubble
Time (days)
Fig. 1 Acropora aspera. Mean values (n = 3–6; ±95% CI) of acetylene reduction (nmol cm–2 h–1) from dead coral surfaces (grey), a summer, and b winter compared to live corals (black) and coral rubble (white stripe). X-axis equals time in days after coral mortality (Ø to 21), except for 11 M sample in summer experiments which was taken 11 months after coral mortality. Light intensity in experimental chambers was maintained at [1,000 lmol cm–2 s–1. Temperature in experimental aquarium was maintained at 28–29C for summer and at 22–23C for winter experiments
Conversion ratio used 3.0b
Summer Live (0.48–0.98)
0.71
0.82
Dead Coral Biofilm (1.08–26.66)
9.19
13.79
0.79 6.96
0.89 7.97
Winter Live (0.30 to 1.16) Dead coral biofilm (1.48–19.07) a
Larkum et al. (1988)
b
Hardy et al. (1968)
123
Acetylene Reduction (nmol cm-2 hr-1)
Coral Reefs (2008) 27:227–236 40
Acetylene Reduction (nmol cm-2 hr-1)
232
40
(a)
(a)
0.8 0.7
30 D12
0.6
% C (org)
20 D8 10 D4 0
D21
C
0.5 0.4 0.3
D0
0.2 0.1
(b)
0.0
30
(b) 0.30
20
0.25 D12 10
D21 0
2
CD 4
%N
0
0.20
D4 0
6 δ15N (‰)
8
10
Fig. 2 Acropora aspera. Mean values (n = 3–6; ±95% CI) of d15N (%; X-axis) versus acetylene reduction (nmol cm–2 h–1; Y-axis) of coral surface algal-bacterial biofilms. DØ to D21 equal time in days after coral mortality, and C equals results from live corals, during a summer, and b winter experiments
Carbon and nitrogen characteristics of the developing biofilms on dead corals are summarised in Fig. 3. Organic C content (Fig. 3a) ranged from 0.20% at day 0 and increased to 0.77 % at day 21. Significant differences were observed for summer experiments over time (F = 101.94, p \ 0.001) and post-hoc analysis revealed that experimental days 14 and 21 were statistically different from the other experimental days and also from each other. Total nitrogen content (Fig. 3b) ranged from 0.03% on day 0 and increased to 0.118% by day 21. A significant difference was observed over time (F = 6.42, p \ 0.05) and post-hoc analysis revealed day 21 was different to days 0 and 8. C:N ratios (Fig. 3c) ranged from 6.54 (mol) at day 0 and fluctuated over the experimental period between 4.26 and 6.12 (mol). Significant differences were observed over time (F = 5.51, p \ 0.05) and post-hoc analysis revealed day 12 was different to day 21.
Discussion Within our current understanding of coral reef nutrient budgets, the role of N2-fixation as a supplier of ‘new’ N is fundamental. N2-fixation is responsible for a considerable
123
0.15 0.10 0.05 0.00
(c)
10
8
C:N
D8
6
4
2
0 0
4
8
12
14
21
Time (Days) Fig. 3 Acropora aspera. Mean (n = 3; ± 95% CI) carbon and nitrogen characteristics for dead coral biofilms from summer (filled circle) and winter (open circle); a percentage of organic carbon; b percentage of nitrogen; c calculated C:N ratios. X-axis equals time in days after coral mortality (Ø to 21)
input of N on coral reefs (Crossland and Barnes 1976; Wilkinson et al. 1984; Larkum et al. 1988; Capone et al. 1992; Shashar et al. 1994a, 1994b; Lesser et al. 2004). The tight coupling and secondary recycling within the reef (e.g., Grigg et al. 1984; Johnstone et al. 1990), and export of nitrogenous compounds from the reef (Johannes et al.
Coral Reefs (2008) 27:227–236
233
Table 2 Acetylene reduction (nmol cm–2 h–1: mean ± SD) and inferred nitrogen fixation rates (mg N2 m–2 day–1) from various global coral reef ecosystems Acetylene reduction
Nitrogen fixation
Substrate studied
Location of study
Reference
3.0–9.9
2.9–9.65
Control
4.68–30.79
Coral skeleton following COTS mortality (various ages)
One Tree Island, Great Barrier Reef
Larkum (1988)
4.8–31.6 3.7–16.4
3.61–15.98
Coral rubble
0.45–1.1
0.44–1.07
Algal turf
2.24–4.38
Outer reef
One Tree Island, Great Barrier Reef
Larkum et al. (1988 )
2.3–4.5 11.0
10.72
Limestone
0.35–0.83
0.34–0.81
Sand
19.52 ± 17.5
21.87 ± 19.61
Sand
Eilat, Red Sea
Shashar et al. (1994a)
55.54 ± 28.5
62.23 ± 31.94
Dead coral surface
9.25 ± 0.5
10.36 ± 0.56
Algal substrate
3.95 ± 4.4
4.43 ± 4.93
Live coral
8.70 ± 7.3
9.75 ± 8.18
Acropora sp.
Eilat, Red Sea
Shashar et al. (1994b)
6.43 ± 2.4
7.21 ± 2.69
Stylophora pistillata
0.61 ± 0.4 7.30 ± 10.75
0.68 ± 0.45 8.18 ± 12.05
Pocillopora damicornis Algal turf
0.33–1.30
0.49–4.9
Sand
1.83 ± 0.26
2.12 ± 0.30
Limestone
1972; Webb et al. 1975) all contribute to reduce the net flux of N into the water column. Despite this, there is little information available on the potential for changes in N2fixation following widespread coral death or the ramifications this may have for coral reef nutrient dynamics. The results of the current study indicate that the relative importance N2-fixation may increase considerably on reefs as a result of coral mortality under thermal stress. Mass coral mortality has occurred in the past (Wilkinson 2000, 2002, 2004) and is predicted to occur with greater frequency in the future (Hoegh-Guldberg 1999). In some instances this mortality has covered 100% of a given reef (Wilkinson 2002). In this context the data presented here reflects the potential for a significant increase in the amount of N entering coral reef ecosystems subject to coral mortality. Furthermore, under increased mortality of corals the usual process of rapid recycling of nutrients may be altered since corals themselves play a dominant role in nutrient conservation (Muscatine and Porter 1977). The increase in N2-fixation on coral surfaces post mortality may increase stocks of available N in the water column. If this does occur it may be sufficient to stimulate local phytoplankton blooms (Furnas et al. 2005), which in turn have the potential to alter sediment nutrient processes through greater deposition of both organic matter and nutrients (Wild et al. 2004; Alongi et al. 2006). Depending on the magnitude of the delivery N and the efficiency of recycling processes within the reef post-mortality, excess N may be exported from the reef (Webb et al. 1975), where it could
US Virgin Island
Williams and Carpenter 1997
Tuamotu Atoll, French Polynesia
Charpy-Roubaud et al. 2001
stimulate local phytoplankton blooms. Although the data from this study shows a significant increase in N2-fixation on coral surfaces for a prolonged period following mortality, the magnitude of the delivery of DIN to areas remote of these surfaces remains to be defined and as such the extent to which reef processes may be altered needs further investigation. N2-fixation rates from live corals and coral rubble from Heron Island were within the range of literature values (Table 2). Subtle differences in the rates may be explained by different physical, biological and chemical differences at individual reefs. Heron Island is at the southern end of the Great Barrier Reef and is therefore subject to wider extremes of temperature and light, both within and between seasons, than reefs closer to the tropics. Differences may also be attributed to the microbial communities inhabiting the coral surface as these populations can vary greatly over small spatial scales due to variations in environmental conditions (Guppy and Bythell 2006; Davey 2006). Rates of N2-fixation from dead corals were up to 30 times higher than those from live corals and coral rubble (Fig. 1). This represents a marked change in N2-fixation on coral skeletons from pre- to post-mortality and, by corollary, a major contribution of ‘new’ N relative to live corals and coral rubble at Heron Island. Moreover, rates obtained 11 months after coral mortality suggest that significantly higher N2-fixation can occur for prolonged periods following the initial death of corals. While, the rates observed from dead corals are comparable with some other studies
123
234
(Larkum 1988; Shashar et al 1994b), it is the levels and duration observed in the context of the magnitude and spatial scales of past bleaching-related mortality (Wilkinson 2000, 2002, 2004) that makes these results important. Although the present study focussed on heat induced coral bleaching as a mechanism of coral mortality, coral mortality can occur via myriad of mechanisms that are not limited to season. For example, cold thermal mortality of corals has been reported and has accounted for coral mortality on some reefs (Hoegh-Guldberg et al. 2005). Beacuse environmental conditions, particularly light and temperature (Larkum 1988) can influence rates of N2-fixation it is important to account for seasonal changes. Rates of N2-fixation rates from dead corals in winter (Fig. 1b) were consistently and significantly higher than rates observed from live corals and coral rubble. However the winter rates from dead corals were lower than rates observed after summer mortality. Nevertheless, the contribution of ‘new’ N from summer mortality coral will be augmented by that from winter mortality events. The considerable increase in N2-fixation, may be explained by marked changes in bacterial communities inhabiting the coral surfaces pre- and post-mortality. An examination of these communities and their structure has shown that they vary considerably spatially and temporally (Davey 2006). Both the abundance and diversity of bacterial phylotypes can increase considerably postperturbation (Mitchell and Cuet 1975; Davey 2006). This change in community composition is likely to correspond to the spatial and temporal changes observed in N2-fixation rates and activity (Paerl 1998). An increase in organic matter as a result of the death and decay of coral tissue may have also directly promoted N2-fixation by providing an energy source and facilitating the formation of oxygen reduced microzones (Paerl et al. 1987). In addition, the increase in algal biomass (Fig. 3) on the coral surfaces post-mortality may have also facilitated N2ixation by creating more oxygen reduced microzones with favourable biogeochemical conditions (Paerl and Prufert 1987). Isotopic analysis of surface biofilms allowed further inferences to be made regarding carbon and nitrogen cycling. Results showed a decrease in d15N values concomitant with an increase in acetylene reduction rates (Fig. 2). Besides confirming nitrogen fixation patterns (Gu and Alexander 1993), this suggests that the ‘new’ nitrogen resulting from nitrogen fixation may either be used by primary producers within the biofilms or is released to the water column (Heikoop et al. 2000). Jones and Stewart (1969) suggested a large fraction of nitrogen fixed by the cyanobacteria Calothrix was transported to plants within the same algal turfs. While, Williams and Carpenter (1998) suggested the products of nitrogen fixed by reef algal turfs
123
Coral Reefs (2008) 27:227–236
were readily transferred to the water column, particularly under increased water motion. The fate of this ‘new’ nitrogen in the longer term remains uncertain in the context of current experimental work. However, knowledge on nutrient effects, sources and sinks in coral reefs may provide an insight into its potential path. Single nutrient pulses to coral reef macroalgal species can produce a substantial increase in biomass of up to 30% (Schaffelke 1999). It is hypothesised here that the large input of nitrogen that occurred in the first 2 weeks after coral death (Fig. 1) may represent such a pulse, particularly considering the spatial scales over which bleaching can occur. Furthermore, the significantly higher nitrogen fixation rates that occur for prolonged periods following coral death (Fig. 1) may represent a ‘‘new’ autochthonous source of nitrogen on reefs subject to bleaching mortality. Without N budgetary data extent to which this allochthonous nutrient augments N inputs remains unclear but may be substantial. Whether this is sufficient to reinforce or push the ecosystem into a phase change as discussed by Done (1992) is unclear but warrants further investigation. Clearly, the fate and form of N derived from these biofilm communities is yet to be defined and there are many aspects that warrant consideration, such as consumption, recycling and influence of grazers (e.g., Williams and Carpenter 1997). In summary, the results from this study highlight the potential for coral bleaching to influence the function of coral reef ecosystems beyond areas such as species loss, alteration in biodiversity and the organismic level trophodymanics that may result. Critically this paper highlights the potential for a substantial change in both the sources and dynamics within the nitrogen budget reef ecosystems subject to large scale coral death. This warrants consideration in future attempts to manage nutrient loads to coral reefs. Acknowledgments MD would like to thank J. Davy, P. Fisher. A. Grinham and the management and staff of Heron Island Research Station for their assistance during the project. MD especially thanks G. Kerven and R. Diocares for analytical assistance, as well as O. Hoegh-Guldberg, S.K. Davy, I. Tibbetts and O. Pantos for reviewing the manuscript. The authors would also like to acknowledge the financial assistance of the Cooperative Research Centre for Reef Research and the Global Education Fund-World Bank Coral Bleaching Program.
References Alongi DM, Pfitzner J, Trott LA (2006) Deposition and cycling of carbon and nitrogen in carbonate mud of the lagoons of Arlington and Sudbury Reefs, Great Barrier Reef. Coral Reefs 25:123–143 Banner AH (1974) Kaneohe Bay, Hawaii: urban pollution and a coral reef ecosystem. Proc 2nd Int Coral Reef Symp 1:685–702
Coral Reefs (2008) 27:227–236 Bianchi TE, Engelhaupt E, Westman P, Andren T, Rolff C, Elmgren W (2000) Cyanobacterial blooms in the Baltic Sea: natural or human induced? Limnol Oceanogr 45:716–726 Breitbarth E, Mills MM, Friedrichs G, LaRoche J (2004) The Bunsen Gas solubility coefficient of ethylene as a function of temperature and salinity and its importance for Nitrogen fixation assays. Limnol Oceanogr Methods 2:282–288 Brodie J, Furnas MJ (1996) Cyclones, river flood plumes and natural water quality extremes in the central Great Barrier Reef. In: Eyles AG, Hunter HM, Rayment GE (eds) Proceeding of the Conference on Downstream Effects of Land Use, Rockhampton. Queensland Department of Primary Industries, Brisbane, pp 367–374 Bunn SE, Loneragan NR, Kempster MA (1995) Effects of acid washing on stable isotope ratios of C and N in Penaeid shrimp and seagrass: implications for food web studies using multiple stable isotopes. Limnol Oceanogr 40:622–625 Capone D, Dunham SE, Horrigan SG, Duguay LE (1992) Microbial nitrogen transformations in unconsolidated coral reef sediments. Mar Ecol Prog Ser 80:75–88 Charpy-Roubaud C, Charpy L, Sarazin G (1996) Diffusional nutrient fluxes at the sediment-water interface and organic matter mineralization in an atoll lagoon (Tikehau, Tuamotu Archipelago, French Polynesia). Mar Ecol Prog Ser 132:181–190 Charpy-Roubaud C, Charpy L, Larkum AWD (2001) Atmospheric dinitrogen fixation by benthic communities of Tikehau Lagoon (Tuamotu Archipelago, French Polynesia) and its contribution to benthic primary production. Mar Biol 139:991–997 Clapcott JE, Bunn SE (2003) Can C4 plants contribute to aquatic food webs of subtropical streams. Freshw Biol 48:1105–1116 Costa OS, Leao Z, Nimmo M, Attrill MJ (2000) Nutrification impacts on coral reefs from northern Bahia, Brazil. Hydrobiologia 440:307–315 Crossland CJ, Barnes DJ (1976) Acetylene reduction by coral skeletons. Limnol Oceanogr 1:153–156 Davey AM (2006) Changes in bacterial communities, carbon and nitrogen dynamics on coral surfaces following mortality: potential implications for reef systems. Ph.D. thesis, University of Queensland, p 241 Done TJ (1992) Phase shifts in coral reef communities and their ecological significance. Hydrobiologia 247:121–132 Dove S (2004) Scleractinian corals with photoprotective host pigments are hypersensitive to thermal bleaching. Mar Ecol Prog Ser 272:99–116 Furnas MJ, Mitchell M, Skuza M, Brodie J (2005) In the other 90%: phytoplankton response to enhanced nutrient availability in the Great Barrier Reef Lagoon. Mar Pollut Bull 51:253–265 Gourlay MR, Colleter G (2005) Wave-generated flow on coral reefs – an analysis for two-dimensional horizontal reef-tops with steep faces. Coast Eng 52:353–387 Grigg RW, Polovina JJ, Atkinson MJ (1984) Model of a coral reef ecosystem III. Resource limitation, community regulation, fisheries yield and resource management. Coral Reefs 3:23–27 Gu B, Alexander V (1993) Estimation of N2 fixation based on differences in the natural abundance of 15N among freshwater N2-fixing and non N2-fixing algae. Oecologia 96:43–48 Guppy R, Bythell JC (2006) Environmental effects on the surface mucus layer bacterial diversity of the reef coral Montastraea faveolata. Mar Ecol Prog Ser 328:133–142 Hardy RWF, Holsten RD, Jackson EK, Burns RC (1968) The acetylene-ethylene assay for N2 fixation: laboratory and field evaluation. Plant Physiol 48:1185–1207 Heikoop JM, Risk MJ, Lazier AV, Edinger EN, Jompa J, Limmon GV, Dunn JJ, Browne DR, Schwarcz HP (2000) Nitrogen-15 signals of anthropogenic nutrient loading in reef corals. Mar Pollut Bull 40:628–636
235 Hoegh-Goldberg O (1999) Climate change, coral bleaching and the future of the world’s coral reefs. Mar Freshw Res 50:839–866 Hoegh-Guldberg O, Fine M, Skirving W, Johnstone R, Dove S, Strong A (2005) Coral bleaching following wintry weather. Limnol Oceanogr 50:265–271 Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K, Johnson CA (2001) Climate change 2001: the scientific basis. Cambridge University Press, Cambridge Hunter CL, Evans CW (1995) Coral reefs in Kaneohe Bay, Hawaii: two centuries of western influence and two decades of data. Bull Mar Sci 57:501–515 Johannes RE, Alberts J, D’Elia CF, Kinzie RA, Pomeroy LR, Sottile W, Wiebe WJ, Marsh JR, Helfrich P, Maqragos J, Meyer J, Smith S, Crabtree D, Roth A, McCloskey LR, Betzer S, Marshall N, Pilson MEQ, Telek G, Clutter RI, DuPaul WD, Webb KL, Wells JM (1972) The metabolism of some coral reef communities: a team study of nutrient and energy flux at Eniwetok. Bioscience 22:541–543 Johnstone RW, Koop K, Larkum AWD (1989) Fluxes of inorganic nitrogen between sediments and water in a coral reef lagoon. Proceedings Limnological Society of New South Wales 110:219–227 Johnstone RW, Koop K, Larkum AWD (1990) Physical aspects of coral reef lagoon sediments in relation to detritus processing and primary productivity. Mar Ecol Prog Ser 66:273–283 Jones K, Stewart WDP (1969) Nitrogen turnover in marine and brackish habitats. IV. Uptake of the extracellular products of the nitrogen-fixing alga Calothrix scopulorum. J Mar Biol Assoc UK 49:701–716 Koop K, Booth D, Broadbent A, Brodie J, Bucher D, Capone D, Coll J, Dennison W, Erdmann M, Harrison P, HoeghGuldberg O, Hutchings P, Jones JB, Larkum AWD, O’Neil JO, Steven A, Tentori E, Ward S, Williamson J, Yellowlees D (2001) ENCORE: the effect of nutrient enrichment or coral reefs, synthesis of results and conclusions. Mar Pollut Bull 42:91–120 Larkum AWD (1988) High rates of nitrogen fixation on coral skeletons after predation by the crown of thorns starfish Acanthaster planci. Mar Biol 97:503–506 Larkum AWD, Kennedy IR, Muller WJ (1988) Nitrogen fixation on a coral reef. Mar Biol 98:143–155 Larsson U, Hagstrom A (1982) Fractionated phytoplankton primary production, exudate release and bacterial production in a Baltic eutrophication gradient. Mar Biol 67:57–70 Lesser MP, Mazel CH, Gorbunov MY, Falkowski PG (2004) Discovery of symbiotic nitrogen-fixing Cyanobacteria in corals. Science 305:997–1000 Mather P, Bennett I (1984) A coral reef handbook: a guide to the flora, fauna and geology of Heron Island and adjacent reefs and cays. Australian Coral Reef Society, Brisbane Mitchell R, Cuet I (1975) Bacterial attack of coral in polluted water. Microb Ecol 2:227–233 Muscatine L, Porter JW (1977) Reef corals: mutualistic symbioses adapted to nutrient-poor environments. BioSci 27:454–460 Paerl HW (1998) Structure and function of anthropogenically altered microbial communities on coastal waters. Curr Opin Biotechnol 1:296–302 Paerl HW, Prufert LE (1987) Oxygen-poor microzones as potential sites of microbial N2- fixation in nitrogen-depleted aerobic marine waters. Appl Env Micro 53:1078–1087 Paerl HW, Crocker KM, Prufert LME (1987) Limitation of N2fixation in coastal marine waters: relative importance of molybdenum, iron, phosphorus, and organic matter availability. Limnol Oceanogr 32:525–536 Postgate JR (1982) The fundamentals of Nitrogen fixation. Cambridge University Press, Cambridge
123
236 Ritchie KB, Smith GW (2004) Microbial communities of coral surface mucopolysaccharide layers. In: Rosenberg E, Loya Y (eds) Coral health and disease. Springer-Verlag, New York, pp 259–263 Ryther JH (1969) Photosynthesis and fish production in the sea. Science 166:72–76 Schaffelke B (1999) Short-term nutrient pulses as tools to assess responses of coral reef macroalgae to enhanced nutrient availability. Mar Ecol Prog Ser 182:305–310 Schaffelke B, Klump DW (1998) Short-term nutrient pulses enhance growth and photosynthesis of the coral reef macroalga Sargassum baccularia. Mar Ecol Prog Ser 170:96–105 Shashar N, Cohen Y, Loya Y, Sar N (1994a) Nitrogen fixation (acetylene reduction) in stony corals: evidence for coral-bacteria interactions. Mar Ecol Prog Ser 111:259–264 Shashar N, Feldstein T, Cohen Y, Loya Y (1994b) Nitrogen fixation (Acetylene-Reduction) on a coral reef. Coral Reefs 13:171–174 Stewart WDP, Fitzgerald GP, Burris RH (1967) In situ studies on N2 fixation using the acetylene reduction technique. Proc Natl Acad Sci USA 58:2017–2078 Webb KL, DuPaul WD, Wiebe WJ, Sottile W, Johannes RE (1975) Enewetok Atoll: aspects of the nitrogen cycle on a coral reef. Limnol Oceanogr 20:198–210 Wild C, Huettel M, Klueter A, Kremb SG, Rasheed MYM, Jorgensen BB (2004) Coral mucus functions as an energy carrier and particle trap in the reef ecosystem. Nature 428:66–70 Wilkinson CR (ed) (2000) Status of the coral reefs of the world: 2000. Australian Institute of Marine Sciences, Townsville
123
Coral Reefs (2008) 27:227–236 Wilkinson CR (ed) (2002) Status of the coral reefs of the world: 2002. Australian Institute of Marine Sciences, Townsville Wilkinson CR (ed) (2004) Status of the coral reefs of the world: 2004. Australian Institute of Marine Sciences, Townsville Wilkinson CR, Williams DM, Sammarco PW, Hogg RW, Trott LA (1984) Rates of nitrogen fixation on coral reefs across the continental shelf of the central Great Barrier Reef. Mar Biol 80:255–262 Wilkinson CR, Linden O, Cesar H, Hodgson G, Rubens J, Strong AE (1999) Ecological and socioeconomic impacts of the 1998 coral mortality in the Indian Ocean: an ENSO impact and a warning of future change. Ambio 28:188–196 Williams SL, Carpenter RC (1988) Nitrogen-limited primary productivity of coral reef algal turfs: potential contribution of ammonium excreted by Diadema antillarum. Mar Ecol Prog Ser 47:145–152 Williams SL, Carpenter RC (1997) Grazing effects on nitrogen fixation in coral reef algal turfs. Mar Biol 130:223–231 Williams SL, Carpenter RC (1998) Effects of unidirectional and oscillatory water flow on nitrogen fixation (acetylene reduction) in coral reef algal turfs, Kaneohe Bay, Hawaii. J Exp Mar Biol Ecol 226:293–316 Yamamuro M, Kayanne H (1995) Rapid direct determination of organic carbon and nitrogen in carbonate-bearing sediments with a Yanaco MT-5 CHN analyser. Limnol Oceanogr 40:1001–1005
