Coral Tissue Thickness as a Bioindicator of Mine-Related Turbidity Stress on Coral Reefs at Lihir Island, Papua New Guinea

USAGE Permission is granted to copy this article for use in teaching and research. Republication, systematic reproduction, or collective redistribution of any portion of this article by photocopy machine, reposting, or other means is permitted only with the approval of The Oceanography Society. Send all correspondence to: info@tos.org or The Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA. Oceanography THE OffICIAL MAGAzINE Of THE OCEANOGRAPHY SOCIETY

coral reef systems before it is possible to identify changes due to unnatural stressors, such as the mining-related sediment loading on Lihir Island (Papua New Guinea, PNG) discussed here.

SedIMeNT-ReL aTed STReSS oN CoR aL ReeFS
Increased turbidity, consequent reduced light levels, and sediment accumulation, are among the most common anthropogenic stressors on coral reefs (see literature reviews in Rotmann, 2004, andFabricius, 2005).These sedimentrelated impacts may affect coral colonies physically through smothering, in the case of high sediment accumulation (e.g., Fabricius, 2005), or physiologically, through a reduction in carbohydrate production and modified oxygen production due to reduced light or photosynthetically available radiation (PAR; e.g., Piniak and Storlazzi, 2008).
The issue becomes more complex when assessing mining-derived, rather than naturally occurring, sediment in the water column as it involves additional contaminants and often greater rates of (chronic) sedimentation.
We found turbidity to be the prominent sediment-related impact on shallowwater communities on Lihir Island.
Turbidity reduces exposure of photosynthetic reef organisms to PAR by absorbing and scattering light (Thomas et al., 2003).Increased turbidity may result in decreased coral species diversity and abundance, changes in coral morphology and metabolism, reduced tissue biomass and lipids, and a shift in coral zonation toward shallower depths (for a review, see Fabricius, 2005).It was also found to lead to both increased and decreased coral bleaching incidents (see, e.g., Rogers, 1979, andAyoub, 2009, respectively).
Corals adapt to increased turbidity by photo-adaptive mechanisms similar to those used to adapt to increased depth, including undergoing changes at the cellular level and changing their symbionts, behaviors, and morphologies (Meesters et al., 2002).Some studies found no coral tissue damage or mortality as a result of elevated turbidity (see Fabricius, 2005), and increased turbidity can be beneficial by supplying organic material as a food source (Anthony, 2000;Weber et al., 2006) or by mitigating the deleterious effects of bleaching by reducing the amount of solar radiation reaching the zooxanthellae (Baker et al., 2008;Sea Rotmann (drsea@orcon.net.nz)

INTRoduCTIoN
Coral reefs are among the most diverse and productive ecosystems on Earth.
Though they are well adapted to surviving natural environmental changes, and acute natural disturbances are critical to maintaining diversity in coral reefs (Richmond, 1993), much concern has been raised in the past two decades about the rapid degradation of coral reefs.Radiation stress (temperature and UV light, together with extreme weather events related to climate change) is the primary stress factor for coral reefs, with cumulative anthropogenic stressors (sedimentation and eutrophication) acting as significant stress-reinforcing factors (e.g., Carpenter et al., 2008;Maina et al., 2011).Anthropogenic stressors do not always fall neatly into only one of these major stress categories (climate change, sedimentation, and eutrophication), yet they can transform natural disturbances into persistent and chronic problems (e.g., terrestrial runoff mixed with pollutants).Thus, it is important to identify natural environmental variability in aBSTR aCT.Work described here assessed the feasibility of using variations in tissue thickness in massive Porites corals as a bioindicator for mine-related sediment stress.We examined parameters influencing coral tissue thickness, including water depth, location, season, and time period within the lunar month.Coral tissue thickness was observed to grow linearly over the lunar cycle until it dropped abruptly by about 20% after the day of the full moon.Although some relationship was observed between tissue thickness reduction and turbidity, no systematic relationship was found between turbidity zones and light levels.The aim was to develop sampling protocols that minimized the effect of natural variability and maximized the potential use of tissue thickness by mine management as a cheap, reliable, real-time indicator of coral stress response to increased turbidity on Lihir Island, Papua New Guinea.This method could prove particularly useful at remote locations or where a fast assessment of coral stress response (< 1 month) needs to be made.Ayoub, 2009).However, it is expected that species composition will change in highly turbid conditions, as some species are significantly more adept at surviving in highly turbid waters than others (see Fabricius, 2005).
Values of suspended sediment concentration (SSC) around coral reefs vary greatly with natural conditions (turbidity may be directly related to SSC via calibration with field samples, and both terms are used interchangeably from here onward): SSC at reefs in pristine conditions have been measured to be less than 1 to more than 30 mg L -1 (see Thomas, 2003, for a literature review).This variation explains why, despite the abundant literature on the subject, it remains impossible to define environmental threshold levels that separate natural conditions from anthropogenic influence (Kirsch, 1999;Orpin et al., 2004;Wolanski et al., 2008;Browne et al., 2010).In the context of this study, it implies that local impact zones had to be defined based on a baseline study and measurements during mining operations in order to assess the effect of mining activity on fringing coral reefs.

CoR aL TISSue ThICKNeSS aS a BIoINdICaToR FoR SedIMeNT STReSS
Various measurements have been used to assess sediment stress response in corals.
However, these techniques do not always give clear results and often involve expensive and sophisticated equipment, making them of limited use in remote locations or developing countries where 90% of the world's reefs are located.
A promising method for remote locations is to use tissue thickness of Porites corals as a bioindicator for coral "health" (Barnes and Lough, 1992;Cooper et al., 2009).Bioindicators are organisms that change visibly or measurably in response to changes in ecosystem health at the level of individuals, populations, and assemblages, and the best bioindicators respond in known and predictable ways.A good bioindicator species is representative of other species in the ecosystem, is abundant as well as easy to identify and to sample, and shows a graduated response to increasing stress.
Coral tissue thickness of Porites refers to the depth of skeleton occupied by tissue.
It fulfils all the criteria listed above for a good bioindicator and was ranked as a high-priority indicator for use in both long-and short-term reef monitoring programs by Cooper et al. (2009).
Indeed, tissue thickness is found to be an important stress-response indicator in corals that bleach (Loya et al., 2001), corals in areas of high sedimentation (Barnes and Lough, 1999), shaded corals (True, 2004), corals living closest to terrestrial and fluvial runoff (Jupiter et al., 2010), corals in competition with turf algae (Quan-Young and Espinoza-Avalos, 2006), corals in waters that have lower nutrient content (Lough and Barnes, 2000;Cooper, 2008), and estrone-treated corals (Tarrant et al., 2004).Tissue thickness of Porites was also shown to vary naturally with water depth (True, 1995) and over the surface of a colony, being highest on the summits (Barnes and Lough, 1992).Finally, there was no difference in tissue thickness between different species of massive Porites (Barnes and Lough, 1992).In this study, we aimed to assess, in detail, the usefulness of the tissue thickness of Porites corals as a bioindicator for mining-related sediment stress on corals.

General Process of Tissue Growth
Porites corals have perforated skeletons in which vertical skeletal elements have many holes, and thus all skeletal cavities above the last dissepiment are interconnected.A dissepiment is a thin, horizontal skeletal bridge formed between vertical skeletal elements that isolates the coral tissue from the skeleton vacated by the tissue during growth (see Plate 1 in Barnes and Lough, 1992).Thus, the uplift of the lower surface of the tissue (polyp base) and emplacement of new dissepiments must occur at the same time over the entire colony.Barnes and Lough (1993)

Influence of the Lunar Cycle on Tissue Thickness
Lunar cycles are a potential environmental cue for synchronous uplift of the tissue layer, as they are known to trigger spawning and larval release (Harrison et al., 1984).Gorbunov and Falkowski (2002) suggested that photoreceptors would enable corals to sense blue moonlight.Despite this recognition of a probable link between lunar cycles and dissepiment emplacement and fine band formation (e.g., Winter and Sammarco,

Mining operations and Sediment Issues
In 1997, the Lihir Management Company (> 25 mg L -1 ), minor (> 10 mg L -1 , which is twice that of oceanic conditions), and control (< 10 mg L -1 ) and were defined based on 50% exceedence probability values of these TSS thresholds (Figure 2a; for the purpose of this paper, TSS is considered to be comparable to SSC).
The  et al., 2003).Data were calibrated in SSC (mg L -1 ) using field samples.The main zoning features were (a) that an extreme turbidity gradient persisted between the inner harbor (turbidity levels of 100-1,000 mg L -1 ) and the adjacent reefs (turbidity levels in the order of 10 mg L -1 ), and (b) that observed zones remained within pre-operations impact predictions.Detailed analysis of turbidity regimes versus particular events such as strong wind and high rainfall events can be found in Thomas et al. (2003).
Four sediment accumulation sensors were also deployed for up to three months where coral reefs were found closest to the activity zone, and this assessment did not detect significant sediment accumulation over fringing coral reefs (Thomas and Ridd, 2005).
Neither of these studies took the biochemical features of mine-derived sediment into account.SAMS (2010) studied mining-related impacts on the deep-sea (> 850 m) fauna at Lihir Island and found significant differences in abundance of macro-and meiofauna between control and impacted sites.

Light Survey
In parallel to turbidity and sediment accumulation, light levels were measured from June to September 2001 at 6 m depth at the southern and northern end of Kapit Reef (i.e., where corals appear to live under the most severe conditions with regard to SSC levels and where a large gradient in the turbidity regime was observed).Light levels were measured with in-built light sensors added to standard JCU OBS (Ridd and Larcombe, 1994).A receiving diode was used to measure the ambient level through a Teflon diffuser in the photosynthetically active region only (i.e., 400 to 700 nm), as controlled by an optical filter.Light sensors were calibrated using a Li-Cor underwater quantum sensor.measured to the nearest 0.1 mm using a digital caliper.Detailed sampling protocol is described in Rotmann (2004).

Tissue Thickness Measurements
Tagged individual corals were revisited every year for three years.

ReSuLTS aNd dISCuSSIoN Natural Changes in Tissue Thickness over a Lunar Cycle
There was a statistically significant, average decrease of 16% ± 2.8% (0.65 mm) in tissue thickness in shallow-water corals and a significant, average decrease of 18% ± 2.5% (0.71 mm) in deepwater corals on the day following the full moon of March 8, 2001 (Figure 3a).Thus, this study ascertained the time at which the abrupt, ~ 20% reduction in tissue thickness occurred to be the day after the full moon.
The rates of change in coral tissue thickness over the lunar month were also documented; they show an almost linear increase between two consecutive full moons (Figure 3a).Additionally, the tissue thickness of deepwater corals was significantly lower (by about 1 mm) than that of shallow-water corals at any one time.
On Lihir Island, tissue uplift on the day following the full moon was also observed in corals sampled at different times of the year (February vs. May), in different years (2001 vs. 2002), at different study sites in various impact zones, and in shallow and deep water (Figure 3b).
In the Great Barrier Reef (GBR), coral tissue thickness was found to vary temporally with seasons (True, 2004), to decrease from inshore to offshore and along a north-south GBR spatial gradient (Barnes and Lough, 1992), and to decrease inversely with water depth (True, 1995).Massive corals like Porites contain "annual" density bands, like tree rings that can be displayed by X-radiography of skeletal slices taken along a colony's growth axis.It has long been thought that fine-density bands found in massive corals were under lunar control, because 12-13 fine bands are found within an annual density band (Buddemeier and Kinzie, 1976;Barnes and Lough, 1993).This study proved that one major aspect of coral skeletal growth, namely tissue uplift and dissepiment formation (see Barnes and Lough, 1993) takes place every month, abruptly, on the day following the full moon.

Natural Versus anthropogenically Induced Turbidity effect Measured by Tissue Thickness
Corals from the severe impact zone as defined by the baseline study (i.e., Kapit Reef and PutPut Point on Figure 2a), corresponding to the transitional zone as defined by the turbidity observations (Figure 2b), had significantly thinner tissues than corals in all other zones (Figure 2d).To illustrate this relationship, Figure 4 shows that for shallowwater corals as measured in 2001, the tissue thickness is significantly reduced in the transitional zone and not reduced everywhere else (Figure 2c).
Deepwater corals had significantly thinner tissue than shallow-water corals at all sites (on average 0.8 mm, or about 20 to 25% thinner), due to a doubling in depth (from < 10 m to the 14-24 m range; Figure 2d).Sanambiet Island corals (naturally impacted by turbidity; Rotmann, 2004) had significantly thinner tissues (on average 2.74 mm ± 0.2 mm) than those at all other sites (average 3.5 mm ± 0.22 mm), except at the predicted severe impact sites (average 2.61 mm ± 0.1 mm; i.e., the observed transitional zone on Figure 2b and Figure 4).These observations suggest a reduction in tissue thickness of about 20-25 % due to naturally increased turbidity.Barnes and Lough (1999) examined the impact on tissue thickness of massive Porites of an up to 100-fold increase in sedimentation resulting from the construction and operation of a gold mine on Misima Island, PNG, and found that tissue thickness decreased significantly with increasing proximity to the mine site, and that corals in the most affected areas were smothered, then buried, after which they died.Sediment accumulation or burial did not occur on reefs affected by mining sediments on Lihir Island because of strong currents and water movement (Thomas and Ridd, 2005).Corals sampled on Lihir Island were mostly affected by turbidity, which is expected to be a less-extreme stressor than sediment accumulation on living surfaces (Woolfe and Larcombe, 1999).
This study, the first attempt at using tissue thickness to assess coral response to turbidity levels (see Cooper, 2008 Turbidity Gradient effect Shown by Tissue Thickness Measurements  This study therefore shows that tissue thickness measurements can be used to delineate boundaries between severe and transitional zones as observed by Thomas et al. (2003) with a measurable tissue thickness reduction (of 20-25% of 3.5 mm), and in the minor and control impact zone with no measurable reduction of tissue thickness.

Light Survey
Variations of natural surface solar radiation levels were large, with a daily maximum ranging from ~ 4,000 to ~ 200 μmol s -1 m -2 between the sunniest and the cloudiest days, respectively.These massive reductions in surface radiation did not last more than one day over the three-month deployment.
Considerable variations of daily maxima were also found in the underwater light level records at 6 m depth, with a range of ~ 500-10 μmol s -1 m -2 at the southern end of Kapit Reef and ~ 750-30 μmol s -1 m -2 at the northern end of Kapit Reef.
Figure 5 shows the percentage of daily underwater surface light that reached the reef at 6 m depth at each location.
Based on the expected PAR, subsurface irradiance as a function of turbidity, depth, and time (Anthony and Fabricius, 2000), 96% of surface irradiance should reach 6 m depth in water with an SSC of 3 mg L -1 , and 12% in water with an SSC of 10 mg L -1 .In this survey, on average, 14% of the daily surface radiation reached the reef at 6 m depth at the northern end of Kapit Reef versus 5% at the southern end, about 2 km away (Figure 5).These data indicate that during this survey, light levels were reduced at both locations compared to normal conditions (defined by an SSC < 3 mg L -1 ).
If the normal light level was taken with an SSC of 10 mg L -1 , which was the observed level at the southern end when there was no particular event such as strong winds or heavy rainfall, light reduction occurred for about 68% of the There is evidence that linear extension (i.e., growth) is somehow linked with tissue reserves (storing lipids like a camel's hump) and that corals with significant energy deficits may sustain skeletal growth rates in the short term by catabolizing tissue reserves (Anthony, 1999).If this is the case, then minimal values of tissue thickness and cessation of dissepiment formation might be linked with reductions or cessation of coral growth rate under stressful conditions.We examined this possible link on the most affected corals on Lihir Island.
In 2001, corals at the most severely impacted site (Kapit Reef) were found to have an average thickness of 2.1 mm (see Figure 6).An additional monthly decrease of about 20-25% would have reduced tissue levels as low as 1.6 mm, Table 1.Impact of parameters studied using Porites tissue thickness with respect to normal conditions.

Coral
Tissue Thickness as a Bioindicator of Mine-Related Turbidity Stress on Coral Reefs at Lihir Island, Papua New Guinea B y S e a R o T M a N N a N d S é V e R I N e T h o M a S R e G u L a R I S S u e F e aT u R e | M I N e Wa S T e d I S P o S a L I N T h e o C e a N Luise harbor with dumping barge in the foreground.Photo courtesy of Lihir Management Company Oceanography | Vol. 25, No. 4 Oceanography | december 2012 53 developed a conceptual model-the "Townsville model of coral growth"-showing that three growth processes are involved in annual density band formation in massive Porites: 1. Extension of skeletal elements at the outer surface of a colony 2. Thickening of skeletal elements throughout the depth of the tissue layer 3. Periodic and abrupt uplift of the lower margin of the tissue layer, reducing tissue thickness by about 20%, and simultaneous building of new dissepiments at the base of the tissue.

Figure 1 .
Figure 1.Location of Lihir Island and gold mine in Luise harbor.Courtesy NSR Environmental Consultants Pty Ltd

((Figure 2
Figure 2. (a) Tissue thickness sampling sites and predicted impact zones based on NSR environmental Consultants Pty Ltd baseline study.(b) observed impact zones based on suspended sediment concentration (SSC) regimes.(c) observed impact zones based on tissue thickness measurements.(d) observed Porites coral tissue thickness along Lihir east coast.
optical backscatter sensors (OBSs;Ridd, 1992) provided a map of impact zones based on turbidity thresholds similar to those used in the prediction survey by NSR (Figure6ain Thomas

Four
Figure 3. (a) Change in tissue thickness during the lunar month (February-March 2001) at deep and shallow nonimpacted sites at Masahet Island (average of nine colonies per depth).(b) Change in tissue thickness for six sites along the impact gradient (averaged over 10 colonies per site) before and after the full moon of May 26, 2002.

Figure
Figure 4. Qualitative correlation between tissue thickness (2001 survey) and impact zones (as defined by turbidity survey).

Figure 2
Figure 2 compares turbidity measurements, coral tissue thickness, and predicted biological impact zones.The data indicate that coral tissue thickness significantly decreases close to the mine site compared to all other sites (Figure 2c), and in the same place where turbidity was measured to be the greatest.More specifically, based on tissue-thickness measurements, the biologically relevant turbidity impact on massive Porites corals on Lihir Island extended from north of Kapit Reef to PutPut Point to the south over a distance of approximately 4 km (Figure 2c).This zone was about half the length of the zone with severe sediment impacts on corals predicted by the baseline survey (NSR, 1989, and Figure 2a).

Figure 5 .
Figure 5. Percentage of daily underwater surface light that reached the reef at 6 m depth at Kapit North and South.

Figure 6 .
Figure6.Comparison of tissue thickness (averaged over the same eight colonies) one day before and one day after the full moon at an impacted site on Kapit Reef two years apart(March 8, 2001, and  February 18, 2003).
the day after full moon depth 20% decrease when depth doubles Natural turbidity impact 20 to 25% anthropogenic turbidity impact 20 to 25% too low to sustain polyp survival, as it would have resulted in partial mortality due to tissue resorption.At this study site, no monthly tissue uplift could be measured in corals with tissue thickness below 2.2 mm, but the sampled corals also showed no visual tissue lesions or patchy die-off.Thus, corals at the site that was the most affected by minerelated turbidity seemed to have adapted to the lower light levels by decreasing their energy expenditure for growth in order to preserve minimum tissue reserves.In 2003, corals from the same site showed a tissue thickness of ~ 3 mm (i.e., a 1 mm or 50% increase since 2001), and tissue uplift was observed following the full moon (Figure 6).These observations could also explain why the main coral species present in the severely impacted sites on Lihir Island is Porites (personal observation of author Rotmann), as similar adaptive mechanisms are not known for corals with imperforate skeletons that do not exhibit monthly tissue uplift.It is unclear from this study how longPorites corals can stop and start linear extension before they die, but several other studies have found that corals have great capacities for adapting to localized stress events and have good potential for recovery after the stressor is removed (e.g.,Maina et al., 2011).Ongoing monitoring of massivePorites tissue thickness levels at highly turbid sites could provide a real-time indicator of coral stress levels, and shed light on the adaptive abilities of these corals(Cooper et al., 2009).In the absence of smothering and burial, the single most significant impact of sediment over coral reefs is caused by light reduction.
CoNCLuSIoNSTable1summarizes the impact of parameters on Porites tissue thickness studied here.This study is the first to