earth and planetary science - marine...

Earth and Planetary Science Letters 487 (2018) 67–73

Contents lists available at ScienceDirect

Earth and Planetary Science Letters

www.elsevier.com/locate/epsl

New evidence for “far-field” Holocene sea level oscillations and links

to global climate records

N.D. Leonard a,∗, K.J. Welsh a, T.R. Clark a, Y.-x. Feng a, J.M. Pandolfi b, J.-x. Zhao a

a School of Earth and Environmental Sciences, The University of Queensland, Brisbane, Qld 4072, Australiab Centre for Marine Science, Australian Research Council Centre of Excellence for Coral Reef Studies, School of Biological Sciences, The University of Queensland, Brisbane, Qld 4072, Australia

a r t i c l e i n f o a b s t r a c t

Article history:Received 12 October 2017Received in revised form 5 February 2018Accepted 5 February 2018Available online xxxxEditor: D. Vance

Keywords:Holocene sea levelGreat Barrier Reef, Australiacoral microatollsuranium thorium dating

Rising sea level in the coming century is of significant concern, yet predicting relative sea level change in response to eustatic sea level variability is complex. Potential analogues are provided by the recent geological past but, until recently, many sea level reconstructions have been limited to millennial scale interpretations due to age uncertainties and paucity in proxy derived records. Here we present a sea level history for the tectonically stable “far-field” Great Barrier Reef, Australia, derived from 94 high precision uranium–thorium dates of sub-fossil coral microatolls. Our results provide evidence for at least two periods of relative sea level instability during the Holocene. These sea level oscillations are broadly synchronous with Indo-Pacific negative sea surface temperature anomalies, rapid global cooling events and glacial advances. We propose that the pace and magnitude of these oscillations are suggestive of eustatic/thermosteric processes operating in conjunction with regional climatic controls.

© 2018 Elsevier B.V. All rights reserved.

1. Introduction

Over the coming century sea-level uncertainty will be one of the greatest challenges of projected global climate change (Milne et al., 2009; Haigh et al., 2014). Not only are millions of peo-ple living in coastal communities at threat of being displaced due to potential sea-level rise, but the vulnerabilities of coastal and intertidal ecosystems are still relatively ambiguous (Hamylton et al., 2014). At local to regional scales the immediate and future impacts of sea-level rise will be governed largely by the rate of sea-level change and adaptation thresholds (Hamylton et al., 2014), however understanding the spatial heterogeneity of rela-tive sea level (RSL) response to eustatic sea-level rise is com-plex (Milne et al., 2009). The Holocene offers potential analogues for understanding future sea-level variability, with geographically divergent RSL histories responding to global melt water follow-ing the end of the Last Glacial Maximum (Clark et al., 1978;Pirazzoli and Pluet, 1991; Lambeck et al., 2014). Where the tran-sition from the Last Glacial Maximum to the early Holocene (∼18,000–8000 yr before present; yr BP) is characterized by a rel-atively globally coherent rapid sea-level rise of ∼125 m due to melting of substantial northern hemisphere ice sheets, the mid-Holocene presents more complex and geographically divergent RSL

* Corresponding author.E-mail address: [email protected] (N.D. Leonard).

histories as a result of glacio-hydro-isostatic and water redistribu-tion processes over sub-millennial to millennial timescales (Clark et al., 1978; Pirazzoli and Pluet, 1991; Mitrovica and Milne, 2002;Lambeck et al., 2014). Yet whether sea level has oscillated sig-nificantly at centennial timescales in response to climate pertur-bations during the Holocene is controversial (Fairbridge, 1961;Chappell, 1983; Baker et al., 2005; Woodroffe and Horton, 2005;Sloss et al., 2007; Woodroffe et al., 2012; Lewis et al., 2013;Lambeck et al., 2014). Therefore, precisely dated geological indica-tors in tectonically stable “far-field” regions offer the best potential to develop RSL histories that may shed light on global eustatic sea-level history.

The intraplate position of Australia (tectonic stability) and its location in the far-field makes it an ideal site for paleo-sea level investigations (Lambeck and Nakada, 1990; Lambeck, 2002). Geo-physical models of glacio-hydro-isostatic RSL response place the Australian East Coast within Zone IV of the of post-glacial melt response (Clark et al., 1978). This zone is characterised by a mid-Holocene RSL highstand, representing the cessation of northern hemisphere ice melt (glacio-eustasy), followed by a RSL regres-sion to present levels where ocean mass redistribution and hydro-isostasy dominate the signal (Clark et al., 1978). In agreement with this model Chappell (1983) used coral microatoll data from the Great Barrier Reef (GBR) to demonstrate a mid-Holocene highstand of 1.0–1.5 m above modern levels by ∼6000 yr BP (where present is defined as before 1950), followed by a linear (smooth fall) RSL

https://doi.org/10.1016/j.epsl.2018.02.0080012-821X/© 2018 Elsevier B.V. All rights reserved.

68 N.D. Leonard et al. / Earth and Planetary Science Letters 487 (2018) 67–73

Fig. 1. Map of Queensland, Australia and the Great Barrier Reef showing the loca-tions of Islands and reefs in the present study. Red circles indicate site locations within each region.

regression to modern levels (Chappell, 1983). However, reports of an oscillating or stepped sea-level signal both relative to Australia (Fairbridge, 1961; Baker et al., 2001, 2005; Lewis et al., 2008) and elsewhere globally (Hamanaka et al., 2012) are numerous. Unfor-tunately, until recent advances in both the accuracy and preci-sion of radiometric dating, understanding sub-centennial sea-level variability has not been realistically possible (Lewis et al., 2008;Lambeck et al., 2014).

Here we aim to refine Holocene “far field” sea level based on 94 new high-precision U–Th dates of coral microatolls (Porites spp.) in conjunction with elevation surveys from the Great Barrier Reef (GBR), Australia. We combine our results with previously pub-lished U–Th dated microatoll data from the southern GBR (Leonard et al., 2016) to assess possible synchronicity of RSL oscillations across ∼10 degrees of latitude (Fig. 1; Fig. SM.2; Materials and Methods). Coral microatolls, especially Porites spp., are considered one of the most reliable palaeo-sea level indicators as the upper flat surface of the colony is constrained by the air–sea interface (Murray-Wallace and Woodroffe, 2014), with modern microatoll elevations lying generally within a vertical range of ∼10 cm of mean low water spring (MLWS) tide level on the GBR (Chappell et al., 1983). We therefore report all elevations relative to modern site specific MLWS level with U–Th ages reported as years before present (yr BP), where present is defined as 1950.

2. Materials and methods

2.1. Sample collection and elevation surveys

As part of a multi-faceted project conducted under the Na-tional Environmental Research Program (NERP), numerous islands and coral reefs of the inshore GBR (11◦S to 23◦S) were visited be-tween 2012 and 2014. Reef flats were visited at the lowest tides possible to enable sampling of fossil Porites sp. microatolls for RSL

reconstructions. Corals were deemed to be in situ based on the orientation of corallite growth direction and relationship to the surrounding substrate (i.e. relative position of other fossil microa-tolls and other fossil reef features).

Using only a single type of sea-level indicator that is well con-strained to a predictable level – mean low water spring (MLWS) tide – reduces the uncertainty of interpretation among sites, and negates the need for elevation interpolation required when a va-riety of sea-level indicators are used. As current hydro-and-glacio-isostatic models for the region are based on a limited number of previously published sea-level records that are geographically and chronologically discontinuous, we present our age-elevation data separated into four latitudinal zones ranging between 11◦S–20◦S (<25 km from the mainland) based on relative location of sites to each other, and width of the continental shelf. No correction has been applied for glacial isostatic adjustment (GIA), as, although this would affect the absolute RSL elevation, it has little effect on the relative position of microatolls to each other within one region.

Elevations of microatolls were taken using a Magnum-Proshot 4.7 Laser Level and Apache Lightning 2 receiver and referenced against a timed-still tide level and, where possible, modern liv-ing counterparts. The elevation was taken from the centre of each microatoll along with the coral surface diameter and GPS location (Table A.1). Elevations were calculated using the nearest tide gauge data from Maritime Safety Queensland (MSQ), time adjusted and reduced to elevation relative to present mean low water spring tide (MLWS; semi-diurnal tides) or mean lowest low water (MLLW; diurnal tides) as given by the Australian Bureau of Meteorology (http :/ /www.bom .gov.au /australia /tides). Tidal range data for each site relative to MSL is provided in Table A.2. Although elevation errors between each sample within sites is minimal and a func-tion of the laser level accuracy (±0.001 m/30 m), we acknowledge the uncertainties of deriving absolute elevations from timed-still tide levels. We assigned a vertical error term to measurements of ±18 cm for Haggerstone Island and a conservative error term of ±15 cm for the remaining sites based on the propagation of tidal error correction and tide tie points of the living population of microatolls at sites compared to modern MLWS/MLLW levels. Our previous dating experience indicated that the centre of the microa-tolls generally had lower detrital inclusions and micro-borings than the edges, which greatly improves the uranium–thorium (U–Th) age accuracy. Therefore, samples of each microatoll were collected for dating from the centre of each colony using a hammer and chisel. Sometimes the centre of the colony was more bio-eroded than the edge, or the exact centre unclear. In this case samples were taken from the edge (E – in Table A.1)

2.2. U–Th dating

Samples were prepared for U–Th dating at the Radiogenic Iso-tope Facility, The University of Queensland, using a pre-cleaning treatment described by Leonard et al. (2016). Crushed and ultra-cleaned samples were picked manually under a binocular micro-scope to allow the best aragonite to be selected for dating (i.e. lacking any detritus, alteration or cements). Picked samples were then weighed (15–50 mg), spiked with a 233U–229Th mixed tracer and dissolved in pre-cleaned Teflon beakers with 15.8 N quartz-double-distilled HNO3. Additionally, 6–10 drops of 30% H2O2 was added to the dissolved sample solutions to remove any remain-ing organics and to ensure complete homogenisation of the spike-sample solution. The Teflon beakers were capped, and the solution heated to 90 ◦C on a hotplate overnight to ensure complete diges-tion. The solution was then dried down completely on a hotplate set at 90 ◦C. Following complete drying, samples were re-dissolved using 0.7 ml of 7 N HNO3 and passed through pre-conditioned Bio-Rad AG1X8 anion resin ion-exchange columns to separate U

N.D. Leonard et al. / Earth and Planetary Science Letters 487 (2018) 67–73 69

Fig. 2. A) U–Th dated microatoll elevation data (metres [m] relative to modern mean low water spring tide) for the Great Barrier Reef (GBR), Australia (GOR N – Gore North, GOR – Gore Island, HAG –Haggerstone Island, ALR – Alexandra Reef, HI – High Island, FTZ – Fitzroy Island, St. I – Stone Island, HAY – Hayman Island - this study) and including U–Th dated microatoll data from Leonard et al. (2016) for the Keppel Islands (KEP; southern GBR). Blue shaded and dashed rectangles represent the periods of inferred negative relative sea level (RSL) oscillations. Dashed black line is the geophysical model (ANT 3) of Nakada and Lambeck (1989) from Halifax Bay (GBR). B) Coral derived (Sr/Ca) sea surface temperature (SST) anomaly data from the Indo-Pacific (◦C; Sadler et al., 2016 and references therein). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

from Th. Quadrupole ICP-MS pre-screening of the collected U and Th solutions was conducted, and where necessary, U and Th so-lutions were remixed in appropriate proportions to maximize the signal. After thorough mixing of the U–Th solution, samples were centrifuged at 3500 rpm for 10 min and then measured fully au-tomatically using a Cetac ASX110 auto-sampler on a Nu Plasma multi-collector inductively coupled plasma mass spectrometer (MC ICP-MS) as described in Clark et al. (2014). Sample ages were cal-culated using the decay constants of Cheng et al. (2000) using Isoplot/Ex software (Ludwig, 2003), and corrected for initial detri-tal 230Th using a two-component mixing correction scheme (Clark et al., 2014).

2.3. Statistical analysis

Relative sea level microatoll data from this study, as well as U–Th microatoll data obtained from the Keppel Islands by Leonard et al. (2016), were combined to derive a single sea-level record for the GBR. The mode of RSL fall to present levels was tested by ap-plying linear and Gaussian models (two–four terms employed) to the data points with 95% confidence in Matlab®. The significance of the normality of the residuals from the models was assessed us-ing the correlation coefficient of the probability in PAST statistical programme.

3. Results

Microatoll age elevation data from the present study (Fig. A.3; Table A.1), combined with previously reported U–Th microatoll data from the Keppel Islands (Fig. 2A; Leonard et al., 2016), show that continental island reef flats had developed by 7000–6500 yr BP on the inshore GBR. In the far north and northern GBR mi-croatolls are found at least 0.5 ± 0.15 m above modern MLWS

from ∼7000–5500 yr BP (Fig. 2A; Fig. A.3). After 5500 yr BP emergent reef flat growth ceases abruptly, with evidence that Holocene MLWS had lowered to close to modern levels between 5200–5000 yr BP at Fitzroy Island (−0.2 m) and Hayman Island (∼0.0 m; Fig. 2A).

Comparatively, data from Stone Island (central GBR) indicates a rising RSL from 6900 to 6600 yr BP, at an average rate of ∼1.4 mm yr−1 (Fig. 2A). This rate of RSL rise is similar to a con-temporaneous record from the Keppel Islands of ∼0.5–1.1 mm yr−1

between ∼6900 and 6200 yr BP (Leonard et al., 2016) suggest-ing that either; a) the timing of the RSL highstand is latitu-dinally displaced due to a lag in water mass redistribution; or more likely b) hydro-isostatic adjustment occurred on this wider section of the continental shelf in the early mid-Holocene, with regional variation possibly related to the NE–SW structural lin-eament boundaries across the shelf (Kleypas and Hopley, 1992;Dechnik et al., 2017).

At Alexandra Reef, a mainland fringing reef, substantial microa-toll development at ∼0.5 ±0.15 m above present did not com-mence until after 5000 yr BP (Fig. 2A; Fig. A.3B). This later initi-ation was likely the result of turbid conditions due to reworking of coastal pre-transgressive sediments being unfavourable for sub-stantial coral growth (Larcombe and Woolfe, 1999). A rapid RSL lowering of −0.3 to −0.5 m occurs at this site at ∼4600 yr BP, persisting for at least 400 yr. Whilst there is overlap of microa-toll U–Th dates at the time of the transition from higher to lower RSL this can be explained by both our sampling strategy and the individual coral morphologies at this site (see Fig. A.1). This RSL lowstand is also supported by evidence on the leeward reef flat of Gore Island, where microatolls occur close to the elevation of their modern counterparts (∼+0.08 m) between 4300 and 4000 yr BP and at Fitzroy Island, where microatolls were found to be below their modern counterparts at 4400 yr BP (Fig. 2A). Emergent reef


Fig. 3. Linear and Gaussian models of relative sea level (RSL) on the Great Barrier Reef, Australia, based on U–Th dated coral microatoll age elevation. Probability plots show the normal distribution of the model residuals. Elevation is in metres (m) relative to present mean low water spring tide (MLWS). U–Th ages are given in years before present (yr BP) where present is defined as 1950. Gaussian 4 represents number of terms included in the analysis. Dotted blue lines are 95% confidence bounds, red dashed squares indicate predictive tendency of the residuals of the models indicating poor fit to the data. The correlation coefficient of normal distribution of the residuals is only significant (p = 0.05) for the Gaussian model.

flat shut down in the Keppel Islands is also synchronous with this lowstand period (Fig. 2A; Fig. A.3E). After 4000 yr BP RSL appears to have risen 0.2–0.3 m to 2800 yr BP (Leonard et al., 2016), after which only a limited number of samples are currently available for interpretation.

Linear and Gaussian models of the microatoll data from the present study, combined with previously published data from the Keppel Islands (Leonard et al., 2016), indicate that significance in the normality of residuals (p = 0.05, n = 130; Filliben, 1975) is only achieved with an increase in terms for a Gaussian model of the RSL data (Fig. 3; Table A.3; adjusted r-squared 0.36) com-pared with lower Gaussian functions (Fig. A.4) or a linear model of RSL regression (Fig. 3; Table A.3; adjusted r-squared 0.15). Both the linear and lower term Gaussian functions overestimate mi-croatoll RSL height from 5500–5000 yr BP and 4500–4000 yr BP (negative residuals) and under estimate RSL from 5000–4500 yr BP (Fig. 3; Fig. A.4), supporting an oscillatory mode of RSL for the GBR.

4. Discussion

4.1. Rapid sea level lowering events

In agreement with geophysical models our data indicates a RSL regression from a mid-Holocene highstand to present levels, largely attributable to ocean syphoning (Chappell, 1983; Mitrovica and Milne, 2002). However, the timing of the highstand on the GBR occurs ∼1000–2000 yr earlier than predicted by earlier geophysi-cal models (Fig. 1; Clark et al., 1978; Nakada and Lambeck, 1989). This temporal offset is a function of the initial modelling param-eters used by Nakada and Lambeck (1989) which were based on

evidence derived from radiocarbon age-elevation data (Chappell, 1983) which, when recalibrated to account for both global and lo-cal marine reservoir effects, would be shifted considerably closer to ∼6500–7000 yr BP (Lewis et al., 2008). The timing of the high-stand at ∼7000 yr BP is therefore in agreement with previous data from the GBR (Chappell, 1983; Lewis et al., 2008) and the Aus-tralian East Coast (Sloss et al., 2007), although we find no evidence of RSL >1 m of present in the mid-Holocene. This may be a result of the microatoll samples delineating a MLWS tide level instead of mean sea level (MSL), so that higher values cannot be discounted. Lewis et al. (2008) reported a systematic −0.5 m elevation offset between microatoll RSL reconstructions and those obtained from fixed biological indicators, likely due to site-specific environmental conditions (e.g. wave energy).

More notable, however, is that the relative elevations of the microatolls surveyed in the present study suggest that the post highstand regression was punctuated by rapid sea level lowering events at 5500 and 4600 yr BP, with a third possible lowstand after ∼2800 yr BP (Fig. 1). The RSL lowering event at 5500 yr BP in the central GBR and the Keppel Islands of at least −0.4 m (Leonard et al., 2016) coincides with significant reductions in reef flat progra-dation (Smithers et al., 2006; Perry and Smithers, 2011), as well as a sudden reef “turn-off” in Moreton Bay by 5600 yr BP (Fig. A.3E; Lybolt et al., 2011; Leonard et al., 2013). The second RSL level lowering event of ∼0.2–0.4 m at 4600 yr BP agrees well with the oscillation proposed by Lewis et al. (2008) for the Australian East Coast (Fig. A.3E), and is also synchronous with reef flat “turn-off” in the Keppel Islands (Leonard et al., 2016) and a significant reef initiation “hiatus” in the southern and northern GBR (Perry and Smithers, 2011). Despite variability in the response of individual reefs, the configuration of the combined RSL signal and synchronic-


ity of re-initiation at some sites between these lowstand periods suggests a return to higher levels.

The coherence of the oscillations across an extensive latitudinal range rules out regional neotectonic activity or local reef/coastal dynamics, although local variations in the absolute level of the RSL reconstruction may be affected. The sub-centennial timescale of the oscillations also precludes hydroisostasy/ocean syphoning as the primary drivers, although these factors explain the overall re-gressive trend following the highstand well. We therefore consider that the oscillations detected in our record may be of eustatic, thermosteric or regional climatic origin, or a combination of these factors.

4.2. Links to climate

At a regional scale, the most important modulating climate sys-tem at annual to decadal scales on the GBR is the El Niño Southern Oscillation (ENSO), with La Niña (El Niño) associated with in-creased (decreased) precipitation in the Austral summer. Although the effect of El Niño/La Niña events on sea surface height on the Australian East Coast is not well understood, in the central Pa-cific sea surface height can vary by as much as 0.3–0.4 m due to the varying phases of ENSO (Woodroffe and McLean, 1998). Recent evidence also suggests that during La Niña phases low lat-itude glaciers advance (Francou et al., 2004) and Antarctic glacier melting is greatly reduced (Dutrieux et al., 2014), increasing the potential of terrestrial water storage in the Southern Hemisphere.

Marine based reconstructions of Holocene climate variabil-ity on the GBR are currently restricted to short time windows. Coral derived sea surface temperature (SST-Sr/Ca) reconstructions demonstrate conditions ∼1 ◦C warmer than present at ∼6200 (re-calibrated 14C) and 4700 yr BP, with a suggested increase in salin-ity range (δ18O) associated with amplified seasonal flood events, suggestive of La Niña (Gagan et al., 1998; Roche et al., 2014). A multi-proxy terrestrial record of pluvial conditions in southern Australia also shows a La Niña like mean state of climate, in-ferred from rainfall maximums at ∼5800–5200, ∼4500 and from ∼3500–2700 yr BP (Gliganic et al., 2014). These periods are also synchronous with phases of dampened El Niño events identified at Laguna Pallcacocha, Ecuador (Fig. A.5F; Moy et al., 2002) and to cooling (or contraction) of the Indo-Pacific warm pool (IPWP) in the western Pacific (Abram et al., 2009). Coral derived Sr/Ca SST anomaly data from the Indo-Pacific (McCulloch et al., 1996;Gagan et al., 1998; Montaggioni et al., 2006; Abram et al., 2009;Duprey et al., 2012; DeLong et al., 2013; Roche et al., 2014;Sadler et al., 2016) also demonstrate warmer than present condi-tions ∼6500 yr BP, followed by a transition to cooler temperatures by ∼5500 yr BP (Fig. 2B).

At a broader scale, foraminiferal abundance and δ18O analysis of a deep sea core off the South Australian coast demonstrates distinct marine cooling events (of possibly ∼2 ◦C) at 5800, 4300 and 2700 yr BP (Fig. A.5B; Moros et al., 2009), which are aligned with cold events identified in the EPICA DOME C ice core. A pe-riod of significant SST cooling and ice expansion was also re-ported for the South Atlantic sector of the Southern Ocean be-tween ∼5500–4700 yr BP (Hodell et al., 2001). In the northern hemisphere, evidence from Greenland (GISP 2) suggests rapid cool-ing events of 1.5–2.0 ◦C from 5600–5400 yr BP, 5000–4700 yr BP, and a stepped cooling trend from 2100–1200 yr BP (Alley, 2000). Rapid cooling also occurs in the North Atlantic (Bond Cycles 4, 3 and 2) at 5900, 4200 and 2800 yr BP (Bond et al., 1997), which align with periods of global glacier advances (Fig. A.5; Denton and Karlén, 1973; Mayewski et al., 2004). With modelled projec-tions of future sea-level rise suggesting a 0.2–0.6 m per +1 ◦C (Church et al., 2013), a first order approximation of observed cool-ing events of ∼1 ◦C in the northern and southern hemisphere

during the Holocene may reconcile the RSL oscillations on the GBR. Although marine and terrestrial palaeo-temperature recon-structions within a given study are not indicative of global mean response, the synchronicity of cooling events and the RSL oscil-lations described here for the GBR are noteworthy and require further investigation. If the oscillations presented here for the GBR are of eustatic/global climatic origin, then oscillations with compa-rable chronologies and/or magnitude should be detectable in other “far-field” locations, even if the absolute elevation of RSLs differ due to local response and the indicators used.

4.3. Far-field sea level records

Mid- to late-Holocene sea level oscillations of over 1 m, as first proposed by Fairbridge (1961), have hitherto been dismissed due to the difficulty in reconciling the required ice volume or ther-mosteric contribution for such substantial eustatic changes. The oscillations identified here for the GBR between 0.2–0.4 m are, however, within the bounds of modelled and hind cast response of global sea levels to relatively minor temperature changes (0.9 ±0.8 m/0.2 ◦C) derived from proxy evidence over the past 3000 yr(Kopp et al., 2016).

Sea level oscillations of ∼0.5 m have also been inferred from both individual microatoll morphological features in Indonesia (Meltzner et al., 2017) and China (Yu et al., 2009), although both occur ∼1000 yr prior to our record. In the north west Pacific at Kodakara Island, disconformities (and hiatus) in an uplifted coral reef at ∼5800, ∼4200 and ∼3200 yr BP were reported, with the latter two events associated with sea level oscillations linked to northern hemisphere cooling (Hamanaka et al., 2012). Geomor-phic evidence from the Atlantic coast of South Africa indicates a rapid sea level fall to below present after ∼5500 yr BP and be-tween 4800–4200 yr BP (Compton, 2006), consistent with the GBR record presented here, although of much larger amplitudes. Facies and faunal interpretations on the coast of Bangladesh also record a stepped RSL regression with rapid lowering from 5900–5700 yr BP, and at 5500 yr BP and a further minor regression after 4800 yr BP (Rashid et al., 2013).

Conversely, a continuous far-field RSL record from Kiritimati, derived from microatoll age-elevation data, suggested that sea level had been within of ∼0.25 m of present throughout the Holocene (Woodroffe et al., 2012). However, separating the interior emer-gent microatoll population data from reef flat data greatly affects the continuity of this record, resulting in a RSL history that is not inconsistent with possible oscillations. The elevation of the fos-sil reef flat microatolls close to present sea level between 4700 and 4100 yr BP, and after 2800 yr BP on the island’s exterior, is in line with the timing of the negative sea-level oscillations pro-posed for the GBR. Furthermore, the most elevated (uncorrected) microatolls from the centre of the atoll occur between our low-stand periods, suggesting that populations inside the former retic-ulate lagoon may either have been isolated from oceanic influence or have reduced tidal flushing, leading to coral demise between 4600–4000 yr BP and 2800–2100 yr BP (within dating uncertain-ties). Unfortunately, the age errors (>±500 yr) of the two earli-est reef flat samples between 6000–5000 yr BP make comparisons with our data difficult. Similarly, a recent analysis of global “far-field” sea level proxy data also reported no evidence of eustatic sea-level variability >0.15–0.20 m for the past 6000 yr. However, the vertical uncertainty of the proxies used in this analysis was sometimes several metres (Lambeck et al., 2014) and age error terms were up to ±700 yr (Woodroffe et al., 2012), making in-terpretation at timescales relevant to rapid climate change events problematic.


5. Conclusions

Supporting a model of RSL instability throughout the Holocene, our study is the first to demonstrate coherent rapid RSL oscilla-tions represented across a large geographic range based upon mea-surements of coral microatoll elevations dated with high-precision U–Th techniques. Despite variation in the response of individual coral reefs to RSL lowering events on the GBR throughout the Holocene, synchronicity of responses at 5500 and 4600 yr BP and after 2800 yr BP co-occur with both oscillations reported at other far-field locations, and with independently documented climate shifts. We propose that GBR RSL oscillations are likely the result of ocean-atmosphere climatic perturbations affecting SSTs, as well as sensitive mountain ice-cap and non-polar icesheet water storage balances in both the northern and southern hemispheres.

With recent advancements in the accuracy and precision of geochronological techniques future research efforts should be con-centrated on obtaining high resolution RSL data from other tec-tonically stable far-field locations. Furthermore, reconstructing high resolution palaeoclimate records, especially in the southern hemi-sphere, is imperative to interpreting possible sub-centennial scale associations between Holocene climate and sea level. Establishing links between sea level and climate in the recent geological past, and refining RSL histories with regard to eustatic changes, will ulti-mately improve the models of future climate change scenarios that are essential for coastal planning and management.

Acknowledgements

We thank James Sadler, Hannah Markham, Martina Prazeres, Mauro Lepore, Jack Coates-Marnane, Ian Butler and Entao Liu for assistance in the field and Dr Ai Duc Nguyen for laboratory as-sistance. We thank three anonymous reviewers and the Editor for constructive comments that improved this manuscript. This study was funded by the National Environmental Research Programme Tropical Ecosystems Hub Project 1.3 to J-xZ, JMP, TRC, Y-xF and others, Australian Research Council Linkage, Infrastructure, Equip-ment and Facilities (LIEF) grant (LE0989067 for the MC-ICPMS) to J-xZ, JMP, Y-xF and others, and an Australian Postgraduate Award to NDL. Samples were collected under permit G12/34,979.1

Appendix A. Supplementary material

Supplementary material related to this article can be found on-line at https://doi.org/10.1016/j.epsl.2018.02.008.

References

Abram, N.J., McGregor, H.V., Gagan, M.K., Hantoro, W.S., Suwargadi, B.W., 2009. Os-cillations in the southern extent of the Indo-Pacific Warm Pool during the mid-Holocene. Quat. Sci. Rev. 28, 2794–2803.

Alley, R.B., 2000. The Younger Dryas cold interval as viewed from central Greenland. Quat. Sci. Rev. 19, 213–226.

Baker, R.G.V., Haworth, R.J., Flood, P.G., 2001. Warmer or cooler late Holocene marine palaeoenvironments?: Interpreting southeast Australian and Brazilian sea-level changes using fixed biological indicators and their δ18O composition. Palaeo-geogr. Palaeoclimatol. Palaeoecol. 168, 249–272.

Baker, R.G.V., Haworth, R.J., Flood, P.G., 2005. An oscillating Holocene sea-level? Re-visiting Rottnest island, western Australia, and the fairbridge eustatic hypothesis. J. Coast. Res., 3–14. Special Issue.

Bond, G., Bonani, G., Showers, W., Cheseby, M., Lotti, R., Almasi, P., deMenocal, P., Priore, P., Cullen, H., Hajdas, I., 1997. A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science 278, 1257–1266.

Chappell, J., 1983. Evidence for smoothly falling sea-level relative to North Queens-land, Australia, during the past 6,000 yr. Nature 302, 406–408.

Chappell, J., Chivas, A., Wallensky, E., Polach, H., Aharon, P., 1983. Holocene palaeo-environmental changes, central to north Great Barrier Reef inner zone. BMR J. Aust. Geol. Geophys. 8, 223–235.

Cheng, H., Edwards, R.L., Hoff, J., Gallup, C.D., Richards, D.A., Asmerom, Y., 2000. The half-lives of uranium-234 and thorium-230. Chem. Geol. 169, 17–33.

Church, J.A., Clark, P.U., Cazenave, A., Gregory, J.M., Jevrejeva, S., Levermann, A., Merrifield, M.A., Milne, G.A., Nerem, R.S., Nunn, P.D., Payne, A.J., Pfeffer, W.T., Stammer, D., Unnikrishnan, A.S., 2013. Sea level change. In: Stocker, T.F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.K., Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P.M. (Eds.), Climate Change 2013: The Physical Science Basis Contribu-tion of Working Group I to the Fifth Assessment Report of the Intergovernmen-tal Panel on Climate Change. Cambridge, United Kingdom and New York, NY, USA.

Clark, J.A., Farrell, W.E., Peltier, W.R., 1978. Global changes in postglacial sea level: a numerical calculation. Quat. Res. 9, 265–287.

Clark, T.R., Roff, G., Zhao, J.-x., Feng, Y.-x., Done, T.J., Pandolfi, J.M., 2014. Testing the precision and accuracy of the U–Th chronometer for dating coral mortality events in the last 100 years. Quat. Geochronol. 23, 35–45.

Compton, J.S., 2006. The mid-Holocene sea-level highstand at Bogenfels Pan on the southwest coast of Namibia. Quat. Res. 66, 303–310.

Dechnik, B., Webster, J.M., Webb, G.E., Nothdurft, L., Zhao, J.-x., 2017. Successive phases of Holocene reef flat development: evidence from the mid- to outer Great Barrier Reef. Palaeogeogr. Palaeoclimatol. Palaeoecol. 466, 221–230.

DeLong, K.L., Quinn, T.M., Taylor, F.W., Shen, C.-C., Lin, K., 2013. Improving coral-base paleoclimate reconstructions by replicating 350 years of coral Sr/Ca variations. Palaeogeogr. Palaeoclimatol. Palaeoecol. 373, 6–24.

Denton, G.H., Karlén, W., 1973. Holocene climatic variations—their pattern and pos-sible cause. Quat. Res. 3. 155IN1175-1174IN2205.

Duprey, N., Lazareth, C.E., Corrège, T., Cornec, F.L., Maes, C., Pujol, N., Madeng-Yogo, M., Caquineau, S., Derome, C.S., Cabioch, G., 2012. Early mid-Holocene SST vari-ability and surface–ocean water balance in the southwest Pacific. Paleoceanog-raphy 27.

Dutrieux, P., De Rydt, J., Jenkins, A., Holland, P.R., Ha, H.K., Lee, S.H., Steig, E.J., Ding, Q., Abrahamsen, E.P., Schröder, M., 2014. Strong sensitivity of Pine Island ice-shelf melting to climatic variability. Science 343, 174–178.

Fairbridge, R.W., 1961. Eustatic changes in sea level. Phys. Chem. Earth 4, 99–185.Filliben, J.J., 1975. The probability plot correlation coefficient test for normality.

Technometrics 17, 111–117.Francou, B., Vuille, M., Favier, V., Cáceres, B., 2004. New evidence for an ENSO impact

on low-latitude glaciers: Antizana 15, Andes of Ecuador, 0 28′S. J. Geophys. Res., Atmos. 109.

Gagan, M.K., Ayliffe, L.K., Hopley, D., Cali, J., Mortimer, G., Chappell, J., McCulloch, M.T., Head, M., 1998. Temperature and surface–ocean water balance of the mid-Holocene tropical western Pacific. Science 279, 1014–1018.

Gliganic, L.A., Cohen, T.J., May, J.-H., Jansen, J.D., Nanson, G.C., Dosseto, A., Larsen, J.R., Aubert, M., 2014. Late-Holocene climatic variability indicated by three nat-ural archives in arid southern Australia. Holocene 24, 104–117.

Haigh, I.D., Wahl, T., Rohling, E.J., Price, R.M., Pattiaratchi, C.B., Calafat, F.M., Dan-gendorf, S., 2014. Timescales for detecting a significant acceleration in sea level rise. Nat. Commun. 5.

Hamanaka, N., Kan, H., Yokoyama, Y., Okamoto, T., Nakashima, Y., Kawana, T., 2012. Disturbances with hiatuses in high-latitude coral reef growth during the Holocene: correlation with millennial-scale global climate change. Glob. Planet. Change 80–81, 21–35.

Hamylton, S.M., Leon, J.X., Saunders, M.I., Woodroffe, C.D., 2014. Simulating reef response to sea-level rise at Lizard Island: a geospatial approach. Geomorphol-ogy 222, 151–161.

Hodell, D.A., Kanfoush, S.L., Shemesh, A., Crosta, X., Charles, C.D., Guilderson, T.P., 2001. Abrupt cooling of Antarctic surface waters and sea ice expansion in the South Atlantic sector of the Southern Ocean at 5000 cal yr B.P. Quat. Res. 56, 191–198.

Kleypas, J.A., Hopley, D., 1992. Reef development across a broad continental shelf, southern Great Barrier Reef, Australia. In: Richmond, R.H. (Ed.), Seventh Interna-tional Coral Reef Symposium. University of Guam Press, Guam, pp. 1129–1141.

Kopp, R.E., Kemp, A.C., Bittermann, K., Horton, B.P., Donnelly, J.P., Gehrels, W.R., Hay, C.C., Mitrovica, J.X., Morrow, E.D., Rahmstorf, S., 2016. Temperature-driven global sea-level variability in the Common Era. Proc. Natl. Acad. Sci. 113, E1434–E1441.

Lambeck, K., 2002. Sea level change from mid Holocene to recent time: an Aus-tralian example with global implications. In: Geodynamics Series, vol. 29, pp. 33–50.

Lambeck, K., Nakada, M., 1990. Late Pleistocene and Holocene sea-level change along the Australian coast. Palaeogeogr. Palaeoclimatol. Palaeoecol. 89, 143–176 (Global and Planetary Change Section).

Lambeck, K., Rouby, H., Purcell, A., Sun, Y., Sambridge, M., 2014. Sea level and global ice volumes from the Last Glacial Maximum to the Holocene. Proc. Natl. Acad. Sci. 111, 15296–15303.

Larcombe, P., Woolfe, K.J., 1999. Terrigenous sediments as influences upon Holocene nearshore coral reefs, central Great Barrier Reef, Australia. Aust. J. Earth Sci. 46, 141–154.

Leonard, N.D., Welsh, K.J., Zhao, J.-x., Nothdurft, L.D., Webb, G.E., Major, J., Feng, Y.-x., Price, G.J., 2013. Mid-Holocene sea-level and coral reef demise: U–Th dating of subfossil corals in Moreton Bay, Australia. Holocene 23, 1841–1852.

Leonard, N.D., Zhao, J.-x., Welsh, K.J., Feng, Y.-x., Smithers, S.G., Pandolfi, J.M., Clark, T.R., 2016. Holocene sea level instability in the southern Great Barrier Reef, Aus-tralia: high-precision U–Th dating of fossil microatolls. Coral Reefs 35, 625–639.


Lewis, S.E., Sloss, C.R., Murray-Wallace, C.V., Woodroffe, C.D., Smithers, S.G., 2013. Post-glacial sea-level changes around the Australian margin: a review. Quat. Sci. Rev. 74, 115–138.

Lewis, S.E., Wu, R.A.J., Webster, J.M., Shields, G.A., 2008. Mid-late Holocene sea-level variability in eastern Australia. Terra Nova 20, 74–81.

Ludwig, K.R., 2003. Isoplot/ex, Version 3: A Geochronological Toolkit for Microsoft Excel. Berkeley Geochronology Center Special Publications.

Lybolt, M., Neil, D., Zhao, J., Feng, Y., Yu, K.-F., Pandolfi, J., 2011. Instability in a marginal coral reef: the shift from natural variability to a human-dominated seascape. Front. Ecol. Environ. 9, 154–160.

Mayewski, P.A., Holmgren, K., Lee-Thorp, J., Rosqvist, G., Rack, F., Staubwasser, M., Schneider, R.R., Steig, E.J., Rohling, E.E., Curt Stager, J., Karlén, W., Maasch, K.A., David Meeker, L., Meyerson, E.A., Gasse, F., van Kreveld, S., 2004. Holocene cli-mate variability. Quat. Res. 62, 243–255.

McCulloch, M., Mortimer, G., Esat, T., Xianhua, L., Pillans, B., Chappell, J., 1996. High resolution windows into early Holocene climate: Sr Ca coral records from the Huon Peninsula. Earth Planet. Sci. Lett. 138, 169–178.

Meltzner, A.J., Switzer, A.D., Horton, B.P., Ashe, E., Qiu, Q., Hill, D.F., Bradley, S.L., Kopp, R.E., Hill, E.M., Majewski, J.M., Natawidjaja, D.H., Suwargadi, B.W., 2017. Half-metre sea-level fluctuations on centennial timescales from mid-Holocene corals of Southeast Asia. Nat. Commun. 8, 14387.

Milne, G.A., Gehrels, W.R., Hughes, C.W., Tamisiea, M.E., 2009. Identifying the causes of sea-level change. Nat. Geosci. 2, 471–478.

Mitrovica, J.X., Milne, G.A., 2002. On the origin of late Holocene sea-level highstands within equatorial ocean basins. Quat. Sci. Rev. 21, 2179–2190.

Montaggioni, L., Le Cornec, F., Correge, T., Cabioch, G., 2006. Coral barium/calcium record of mid-Holocene upwelling activity in New Caledonia, South-West Pa-cific. Palaeogeogr. Palaeoclimatol. Palaeoecol. 237, 436–455.

Moros, M., De Deckker, P., Jansen, E., Perner, K., Telford, R.J., 2009. Holocene cli-mate variability in the Southern Ocean recorded in a deep-sea sediment core off South Australia. Quat. Sci. Rev. 28, 1932–1940.

Moy, C.M., Anderson, D.M., Seltzer, G.O., Rodbell, D.T., 2002. Variability of El Niño/Southern Oscillation activity at millennial timescales during the Holocene epoch. Nature 420, 162–165.

Murray-Wallace, C.V., Woodroffe, C.D., 2014. Palaeo-sea-level indicators. In: Qua-ternary Sea-Level Changes: A Global Perspective. Cambridge University Press, Cambridge, New York, pp. 95–98.

Nakada, M., Lambeck, K., 1989. Late Pleistocene and Holocene sea-level change in the Australian region and mantle rheology. Geophys. J. Int. 96, 497–517.

Perry, C., Smithers, S., 2011. Cycles of coral reef ‘turn-on’, rapid growth and ‘turn-off’ over the past 8500 years: a context for understanding modern ecological states and trajectories. Glob. Change Biol. 17, 76–86.

Pirazzoli, P.A., Pluet, J., 1991. World Atlas of Holocene Sea-Level Changes. Elsevier, Amsterdam, New York.

Rashid, T., Suzuki, S., Sato, H., Monsur, M.H., Saha, S.K., 2013. Relative sea-level changes during the Holocene in Bangladesh. J. Asian Earth Sci. 64, 136–150.

Roche, R.C., Perry, C.T., Smithers, S.G., Leng, M.J., Grove, C.A., Sloane, H.J., Unsworth, C.E., 2014. Mid-Holocene sea surface conditions and riverine influence on the inshore Great Barrier Reef. Holocene 24, 885–897.

Sadler, J., Webb, G.E., Leonard, N.D., Nothdurft, L.D., Clark, T.R., 2016. Reef core in-sights into mid-Holocene water temperatures of the southern Great Barrier Reef. Paleoceanography 31, 1395–1408.

Sloss, C.R., Murray-Wallace, C.V., Jones, B.G., 2007. Holocene sea-level change on the southeast coast of Australia: a review. Holocene 17, 999.

Smithers, S.G., Hopley, D., Parnell, K.E., 2006. Fringing and nearshore coral reefs of the Great Barrier Reef: episodic Holocene development and future prospects. J. Coast. Res., 175–187.

Woodroffe, C.D., Horton, B.P., 2005. Holocene sea-level changes in the Indo-Pacific. J. Asian Earth Sci. 25, 29–43.

Woodroffe, C.D., McGregor, H.V., Lambeck, K., Smithers, S.G., Fink, D., 2012. Mid-Pacific microatolls record sea-level stability over the past 5000 yr. Geology 40, 951.

Woodroffe, C.D., McLean, R.F., 1998. Pleistocene morphology and Holocene emer-gence of Christmas (Kiritimati), Island, Pacific Ocean. Coral Reefs 17, 235–248.

Yu, K.-F., Zhao, J.-X., Done, T., Chen, T.-G., 2009. Microatoll record for large century-scale sea-level fluctuations in the mid-Holocene. Quat. Res. 71, 354–360.

earth and planetary science - marine...

Documents