Hoey Et Al. Elizabeth and Middleton Marine Survey 2018 Final

STATUS AND TRENDS FOR SHALLOW REEF HABITATS AND ASSEMBLAGES AT ELIZABETH AND MIDDLETON REEFS, LORD HOWE MARINE PARK

Andrew S. Hoey, Morgan S. Pratchett, Katie Sambrook, Sallyann Gudge, and Deborah J. Pratchett

Produced for The Director of National Parks, June 2018

*Corresponding author – Professor Morgan Pratchett, ARC Centre of Excellence for Coral Reef Studies, James Cook University, Townsville QLD 4811. E-mail: [email protected], Phone/ Fax: (07) 47815747 / 47816722

Page 1

In responding to a tender from the Department of Environment, a team of researchers representing the ARC Centre of Excellence for Coral Reef Studies at James Cook University (JCU) completed surveys of the coral reef fauna at Elizabeth and Middleton Reef. The field team comprised Professor Morgan Pratchett, Dr Andrew Hoey, Ms Katie Sambrook and Mrs Sallyann Gudge. This report was prepared by Professor Morgan Pratchett, Dr Andrew Hoey, Ms Katie Sambrook, Mrs Sallyann Gudge and Mrs Deborah Pratchett On the cover - Photograph taken by Scott Ling in Blue Holes, Middleton Reef on 27 February 2018

© Commonwealth of Australia 2018 This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without prior written permission from the Commonwealth. Requests and inquiries concerning reproduction and rights should be addressed to the Commonwealth Copyright Administration, Attorney General’s Department, Robert Garran Offices, National Circuit, Barton ACT 2600 or posted at http://www.ag.gov.au/cca

The views and opinions expressed in this publication are those of the authors and do not necessarily reflect those of the Australian Government.

This report has been produced for the sole use of the party who requested it. The application or use of this report and of any data or information (including results of experiments, conclusions, and recommendations) contained within it shall be at the sole risk and responsibility of that party. JCU does not provide any warranty or assurance as to the accuracy or suitability of the whole or any part of the report, for any particular purpose or application.

Page 2

1 Executive Summary

Elizabeth and Middleton Reefs are the world’s southernmost open ocean platform reefs, supporting a unique mix of tropical and subtropical species. In-water field surveys were conducted in 2011, 2014 and 2018 using consistent methods at the same locations, enabling rigorous assessment of trends in benthic habitats and associated assemblages of reef fish and mobile macro-invertebrates.

The most recent surveys were conducted from February 25th to March 4th, 2018, with 4 days (25-29 February) spent at Middleton Reef and 4 days (1-4 March) at Elizabeth Reef. Surveys were conducted at a total of 17 sites across both Middleton and Elizabeth reefs. Where possible, surveys were conducted in two distinct habitats (shallow reef crest or top and deep reef slope or base) at each site. Benthic composition and habitat structure, as well as coral health, densities and biomass of functionally important reef fishes, densities of site attached and endemic fishes, and densities of non-coral invertebrates were recorded along four replicate 50 m transects in each habitat at each site.

Main findings include:

• Mean cover of hard (Scleractinian) corals was high in 2018 (Elizabeth Reef: 39.95% ±1.72 SE; Middleton Reef: 26.62% ±1.87 SE) and has increased by >40% since 2014. There was however, low incidence of coral bleaching in 2018, which may have worsened after surveys were completed due to the significant accumulation of heat stress that occurred in 2017/18.

• Aside from bleaching, there was very low incidence of coral predation, disease, and or other unexplained coral injuries. Very few crown-of-thorns starfish or Drupella spp. were recorded in 2018, though there is evidence of a persistent low-density population of crown-of-thorns starfish at Elizabeth Reef. Given their considerable reproductive potential and low visitation to these locations, it might be prudent to remove these starfish so as to minimise the likelihood of future population outbreaks.

• Cover of macroalgae (conspicuous seaweeds >5 mm in height) was high in back reef and lagoonal habitats, but mostly very low in exposed reef habitats.

Page 3

Both the cover and the predominant type of macroalgae found at each site was largely unchanged since 2014. Similarly, the physical structure (topographic complexity) of reef habitats was very consistent among sites and largely unchanged from 2011 to 2018.

• The abundance of herbivorous fishes recorded at Elizabeth and Middleton Reefs in 2018 were similar to those recorded in 2011 and 2014 (Elizabeth: 2011: 67.3 fishes/250m2, 2014: 54.8 fishes/250m2, 2018: 58.8 fishes/250m2; Middleton: 2011: 72.8 fishes/250m2, 2014: 59.6 fishes/250m2, 2018: 60.0 fishes/250m2).

• The abundance of endemic fishes (Amphiprion mccullochi, Chaetodon tricinctus, and Coris bulbifrons) varied greatly among study sites, but have changed very little from 2011 to 2018. However, the abundance of C. tricinctus at Elizabeth Reef has declined from 1.89 fish/100m2 in 2014 to 1.28 fish/100m2 in 2018, despite increases in coral cover during this period.

• The abundance of Galapagos sharks (Carcharinus galapagensis) at Elizabeth Reef in 2018 (10.2 individuals/ha ± 4.0 SE) was higher than recorded in 2014 (6.58 individuals/ha ± 1.69 SE). In contrast, the abundance of C. galapagensis at Middleton in 2018 (8.9 individuals/ha ± 2.4 SE) was lower than recorded in 2014 (17.11 individuals/ha ± 2.99 SE).

• The abundance of black cod (Epinephelus daemelii) at Elizabeth (1.50 fish/ha ± 0.35 SE) and Middleton Reefs (1.41 fish/ha ± 0.55 SE) in 2018, was lower than recorded in 2014 (2.50 fish/ha ± 0.55 SE and 2.24 fish/ha ± 0.53 SE, respectively).

Page 4

2 Recommendations

1. Recurrent monitoring, using consistent and standardised protocols, is fundamental in assessing the status and trends of coral reef habitats and associated fish and non-coral invertebrate assemblages at Elizabeth and Middleton Reefs. Nonetheless, it is very difficult to establish specific causes of any marked changes in coral reef environments when the interval between recurrent surveys is greater than a year, and up to 4-5 years. Importantly, if or when there are major disturbances (e.g., outbreaks of crown-of-thorns starfish or climate-induced coral bleaching) affecting these reefs, the current monitoring regime will be insufficient to effectively assess the specific effects of these important and extremely influential events.

2. Unpredictable and acute disturbances (e.g., outbreaks of crown-of-thorns starfish or climate-induced coral bleaching) represent the greatest and most immediate risk to the status and condition of shallow reef environments and assemblages at Elizabeth and Middleton Reefs. It is critical therefore, that ongoing monitoring is undertaken to assess these risks, and that there is effective capacity to respond in a timely fashion to assess localised impacts and associated changes in reef assemblages. For example, monitoring should be undertaken during, or immediately following, periods of extreme temperatures and elevated bleaching risk (NB. In compiling this report it has become apparent that the risk of bleaching has increased since we undertook surveys in February/ March and the accumulated heat stress this year is among the highest ever recorded for Elizabeth and Middleton Reefs). In the absence of frequent or responsive surveys, the utility of occasional visitors should be maximised, both at Elizabeth and Middleton Reefs. At present there is an obligation for visitors to Elizabeth Reef to report on reef condition (owing to the current management zoning), and this requirement should should continue to be a requirement for all visitors to Elizabeth (proposed TELHIRUZ03) whether undertaking fishing or not, as well as being extended to include visitors to Middleton Reef. Visitor reports (however infrequent) are invaluable to potentially alert managers to any apparent concerns regarding threats to coral and ecosystem health.

Page 5

3. The black cod (Epinephelus daemelii) is listed as a vulnerable species within NSW and is totally protected from any form of fishing. Recent surveys have revealed declines in abundance of E. daemelii at both Elizabeth and Middleton Reefs since 2014, with no E. daemelii being recorded at the lagoon sites on Middleton Reef in 2018. Declines in the abundance of this species are concerning given that there are very limited numbers of E. daemelii remaining elsewhere throughout their geographic range. As such, much more information is needed on the biology and ecology of this species, both at Elizabeth and Middleton Reefs and at other nearby reefs (e.g., Lord Howe Island) to understand the population dynamics, movement patterns and vulnerability of E. daemelii. Moreover, greater efforts may be warranted to highlight the protected status of Black cod to visiting vessels and recreational fishers to assist in reducing incidental by-catch and any fishing related mortality.

4. Recurrent surveys to assess the status and trends in coral health and reef condition are conditional upon effective collaborations between coral biologists and reef ecologists (not necessarily from James Cook University), with managers from Commonwealth (the Director of National Parks) and state agencies (NSW Department of Primary Industries). An important component of these collaborations is to ensure all parties are present and represented during each and every expedition (especially given limited connectivity whilst in the field), to facilitate appropriate responses to emerging issues (e.g., re-directing survey effort to document impacts of specific threats and disturbances). This is especially critical for Elizabeth and Middleton Reefs given the limited frequency of surveys, and potential for large changes to become apparent between successive surveys.

Page 6

Table of Contents

1. Executive Summary 3 2. Recommendations 5 3. Introduction 8 3.1 Elizabeth and Middleton Reefs (Lord Howe Marine Park) 9 3.2 History of Coral Reef Surveys at Elizabeth and Middleton 12 Reefs 3.3 Objectives and scope 14 4. Methods 15 4.1 Coral and reef habitats 15 4.2 Non-coral invertebrates 19 4.3 Coral reef fishes 20 4.4 Data analyses 21 5. Findings 22 5.1 Coral and reef habitats 22 5.2 Non-coral invertebrates 29 5.3 Coral reef fishes 31 6. Conclusions 51 6.1 Coral and reef habitats 51 6.2 Non-coral invertebrates 52 6.3 Coral reef fishes 53 7. Acknowledgements 57 8. References 58

Page 7

3 Introduction

Coral reefs are among the most threatened natural ecosystems globally, owing to a long-history of anthropogenic degradation and exploitation (Jackson et al. 2001; Pandolfi et al. 2003), as well as their extreme vulnerability to global climate change (Hughes et al. 2017a, 2018). Even before recent effects of global climate change (which has caused successive years of severe coral bleaching in many coral reef regions), one-fifth of the world’s coral reefs had been effectively destroyed, meaning that coral cover had been reduced to such low levels (<5%) that these systems were no longer effectively functioning as coral reefs, or that the reefs have been completely removed and/ or buried by mining or land reclamation (Wilkinson 2008). In east Africa, south-east Asia, and the central and southern Caribbean, a disproportionate number of coral reefs have been lost due high levels of sustained human activities, including chronic pollution, eutrophication, sedimentation, overfishing and/or destructive fishing practices (Pandolfi et al. 2003; Wilkinson 2008). Even isolated coral reefs that are well removed from effects of coastal development are being increasingly affected by insidious effects of global fisheries (Graham et al. 2010) and ongoing climate change (Hughes et al. 2003; Hoegh- Guldberg et al. 2007).

Relatively isolated (Gilmour et al. 2013) and high latitude (Beger et al. 2014) coral reef ecosystems are increasingly viewed as important refuges, and considered to be relatively less susceptible to direct anthropogenic pressures and global climate change, respectively. However, when isolated reefs are subject to major disturbances recovery may be highly constrained compared to well-connected reef ecosystems because there is less opportunity for larval replenishment from other unaffected reef areas and populations (Graham et al. 2006; Smith et al. 2008, Gilmour et al. 2013). High-latitude reefs systems may also be more (not less) susceptible to global climate change because they are disproportionately affected by changing seawater chemistry (specifically, ocean acidification and declining aragonite saturation), which may impose increasing constraints on calcification and reef accretion (Anderson et al. 2015), because there is a natural gradient of declining aragonite saturation with increasing latitude. There is however, evidence that corals subject to naturally higher fluctuations in temperature, are more

Page 8

resistant to ocean warming (Oliver and Palumbi 2011), and potentially much less susceptible to coral bleaching. Coral bleaching has been documented on high- latitude reefs in Australia (e.g., Harrison et al. 2011) and Japan (Nishiguchi et al. 2018) though the relevant susceptibility and severity of bleaching (compared to low-latitude reefs) is equivocal.

The foremost management strategy to address increasing anthropogenic pressures on coral reefs (as for other marine and coastal ecosystems) is marine protected areas (MPAs) or marine parks (Graham et al. 2011, Edgar et al. 2014), which limit or prohibit extractive activities within specified areas. The effective implementation (combining both designation and appropriate compliance) of marine parks, and in particular no-take areas, can have significant benefits for abundance and diversity of fisheries-target species, especially in large, long- established and well-managed (well enforced) marine parks (e.g., Russ and Alcala 2010, Edgar et al. 2014). However, given the widespread and accelerating degradation of coral reef ecosystems, marine parks are increasingly expected to provide additional benefits beyond protection of major targeted species (Graham et al. 2011), specifically related to preserving key ecological functions and maximising ecosystem resilience. While there is increasing evidence that marine parks do contribute to enhancing the ecological integrity and resilience of coral reef ecosystems, it is also apparent that this alone may be insufficient to effectively conserve these critically important habitats (Hughes et al. 2017b).

3.1 Elizabeth and Middleton Reefs (Lord Howe Marine Park)

Elizabeth and Middleton Reefs located within the Tasman Sea are the world’s southernmost open ocean platform reefs, and supporting a unique mix of tropical and subtropical species (Choat et al. 2006). These emergent reef systems are contained within the Lord Howe Marine Park (Figure 1). The northernmost zone encompassing Middleton Reef is a Sanctuary Zone, which allows “passive use by the public”, but no fishing. Recreational fishing is permitted in the southern “Habitat Protection Zone” that encompasses Elizabeth Reef, though fishing pressure is presumed to be very low, owing to its isolation.

Page 9

( ( ( ( ( 6 6

)RUPHU(OL]DEHWKDQG 0LGGOHWRQ5HHIV 0DULQH1DWLRQDO 1DWXUH5HVHUYH Middleton reef 6 6

ƍ6

Elizabeth reef ƍ6 6 6

)RUPHU/RUG+RZH ,VODQG0DULQH3DUN QP &RPPRQZHDOWK ZDWHUV

/RUG 6 6

+RZH ,VODQG

ƍ6 0DULQH 3DUN 16:

Lord Howe Island ƍ6

Lord QP Howe Rise 6 6

P %DWK\PHWULFFRQWRXUVDUH VKRZQDWPGHSWKLQWHUYDOV DQGDUHLQGLFDWLYHRQO\

P NP $SSUR[LPDWHVFDOH 6 6

QP Tasman Basin 3URMHFWLRQ*HRJUDSKLFV 'DWXP*'$ HIIHFWLYHO\:*6

127)251$9,*$7,21$/385326(6 Flinders Seamount &RRUGLQDWHV DUH H[SUHVVHG LQ GHJUHHV DQG PLQXWHV :KHUH ERXQGDULHV DUH QRW JD]HWWHG DV ZKROH PLQXWHV WKH FRRUGLQDWHV VKRZQ KDYH EHHQ URXQGHG 6 6

WR GHFLPDO SODFHV RI D PLQXWH PHWUH RI

JD]HWWHGFRRUGLQDWHV

Former Elizabeth and Middleton Reefs Marine National Nature Reserve and Former Lord Howe Island Marine Park (Commonwealth waters)

Zoning Other marine protected area $XVWUDOLDQ*RYHUQPHQWGDWDVRXUFHV 6DQFWXDU\=RQH ,8&1,D 'R(( &ROODERUDWLYH$XVWUDOLDQ3URWHFWHG$UHDV'DWDEDVH 6WDWHPDULQHSDUN 16: *HRVFLHQFH$XVWUDOLD $XVWUDOLDQ0DULWLPH%RXQGDULHV +DELWDW3URWHFWLRQ=RQH ,8&1,, $0% Y

6 *HRVFLHQFH$XVWUDOLD *(2'$7$7232. 6

Maritime boundaries *HRVFLHQFH$XVWUDOLD $XVWUDOLDQ%DWK\PHWU\DQG7RSRJUDSK\ +DELWDW3URWHFWLRQ=RQH ,8&1,9 3URGXFHGE\WKH(QYLURQPHQWDO5HVRXUFHV,QIRUPDWLRQ1HWZRUN (5,1 /LPLWRI$XVWUDOLDQH[FOXVLYH $XVWUDOLDQ*RYHUQPHQW'HSDUWPHQWRIWKH(QYLURQPHQWDQG(QHUJ\ HFRQRPLF]RQH &RPPRQZHDOWKRI$XVWUDOLD

Figure 1: The spatial extent and location of specific management zones within the Lord Howe Marine Park. Notably, Middleton Reef is a Sanctuary Zone (IUCN Ia), while Elizabeth is classed as Habitat Protection Zone (IUCN II).

Page 10

Elizabeth and Middleton Reefs are critically important coral reef environments, owing to their location and isolation. These reefs support a number endemic species or species that are generally rare over most of their geographic range. Most notably, Elizabeth and Middleton Reefs are renowned for being the last remaining stronghold for the black cod (Epinephelus daemelii), which occurs throughout southwestern Pacific, but has been overfished throughout much of its range (Choat et al. 2006). Epinephelus daemelii is listed as vulnerable in New South Wales, and is also listed as a ‘vulnerable’ species under the Environment Protection and Biodiversity Conservation Act 1999.

3.2 History of Coral Reef Surveys at Elizabeth and Middleton Reefs

There have been at least 13 scientific surveys undertaken at Elizabeth and/ or Middleton Reefs since 1979 (Table 1), including this survey and corresponding assessments by Reef Life Survey. Collectively, these surveys represent the longest running and most intensive temporal series of surveys of coral reef habitats across Australian Marine Parks – the 58 marine parks managed by the Director of National Parks and Parks Australia (Hoey and Pratchett 2017). Recurrent surveys of the shallow reef environments have revealed very gradual, but sustained, recovery of coral cover at Elizabeth and Middleton Reefs from 10% cover in 1994 up to 20-30% in 2014 (Hoey and Pratchett 2017). Low levels of coral cover reported in the 1990s are widely attributed to coral depletion caused by a localised outbreak of crown-of-thorns starfish in the 1980s, where high densities of these corallivorous starfishes were documented by Harriot (1998). While the subsequent recovery of coral assemblages is encouraging, it is apparent that rates of coral recovery at these locations are slow, especially compared with isolated reefs at lower latitudes (e.g., Scott Reef, Gilmour et al. 2013). Constraints on rates of recovery are probably due to low levels of coral recruitment and population replenishment (Pratchett et al. 2015), as well as environmental constraints on coral calcification and growth (Anderson et al. 2015). It is possible, however, that recruitment rates will increase in accordance with the size and abundance of adult corals (Gilmour et al. 2013), leading to marked increases in rates of recovery once local coral cover reaches certain thresholds.

Page 11

Table 1. Marine surveys undertaken at Elizabeth and/ or Middleton Reef since 1979. For a comprehensive overview of surveys conducted prior to 1979 see Australian Museum (1992).

No. Year Locations Purpose Reference

1 1979 Elizabeth & Coral diversity Australian Museum Middleton 1992

2 1984 Elizabeth & Coral diversity Australian Museum Middleton 1992

3 1981 Middleton Assess impacts of Harriot 1998 COTS

4 1987 Elizabeth & Biodiversity Australian Museum Middleton assessment 1992

5 1994 Elizabeth & Marine ecological Choat et al. Middleton survey unpublished data

6 2003 Elizabeth Marine ecological Oxley et al. 2004 survey

7 2006 Elizabeth & Marine ecological Choat et al. 2006 Middleton survey

8 2007 Elizabeth & Rapid assessment Hobbs and Feary 2007 Middleton of reef health

9 2011 Elizabeth & Marine ecological Pratchett et al. 2011 Middleton survey

10 2012 Elizabeth & Reef Life Survey Stuart-Smith et al. 2017 Middleton (RLS)

11 2014 Elizabeth & Marine ecological Hoey et al. 2014 Middleton survey

12 2018 Elizabeth & Marine ecological This report Middleton survey

13 2018 Elizabeth & Reef Life Survey Middleton (RLS)

Page 12

3.2 Objectives and scope

The purpose of this study was to assess the status and trends of shallow coral reef ecosystems and organisms at Elizabeth and Middleton Reefs, within the Lord Howe Marine Park, using standardised methods and at fixed locations to allow meaningful comparisons across 2011, 2014 and 2018. Replicate area-based surveys (belt transects) were undertaken to quantify the contemporary abundance of select fish species, as well as ecologically and economically important mobile invertebrates (e.g., holothurians). Coral cover and composition, as well as overall habitat structure, were also quantified along the same transects (n = 4) in two distinct depth habitats in 2011, 2014 and 2018. In 2018, we re-visited and successfully surveyed 16 sites that were surveyed in 2011 and 2014, and also extended surveys to include three new sites at Elizabeth Reef, including additional sites on the exposed reef front where adverse weather has prohibited prior surveys.

These surveys provide rigorous quantitative information on: i) benthic cover and composition, including estimates of percentage cover for hard (Scleractinian) and soft (Alcyonarian) corals, macroalgae, and other sessile organisms ii) structural complexity of reef habitats iii) coral health and injuries caused by coral disease, corallivorous invertebrates (e.g., Acanthaster spp and Drupella spp), or reef fishes iv) abundance of small/ juvenile corals (<5cm diameter), as a proxy of coral recruitment and population replenishment, v) size and abundance of large iconic species, including black cod (Epinephelus daemelii) and Galapagos shark (Carcharinus galapagensis), vi) abundances of other reef fishes including the endemic doubleheader wrasse (Coris bulbifrons), rabbitfishes, drummers, damselfishes and butterflyfishes, and vii) abundance of holothurians, urchins and other ecologically or economically important reef-associated invertebrates.

Page 13

4 Methods

This study was conducted at Elizabeth and Middleton Reefs from February 24th to March 3rd, 2018. Sampling was undertaken at a total of 19 sites across both reefs (Table 2; Figure 2). Three new sites were added at Elizabeth Reef (sites 8, 9, and 10) to provide better representation of exposed reef front habitats, and take advantage of the relatively calm conditions. At each site, surveys were conducted within each of two depth zones, i) the reef crest at 2-4m and ii) the reef slope at 6- 10m. In each depth zone at each site, four replicate 50-m transects were run parallel to the depth contour, with up to 10 metres between successive transects (n = 152 transects in total across both reefs). Along each transect (a) benthic cover and composition, (b) reef topographic complexity, (c) juvenile coral densities (as a proxy for coral replenishment), (d) coral condition/health, (e) herbivorous fish assemblages, (f) endemic and site-attached fish densities, (g) apex predator densities, and (h) mobile invertebrates were quantified using the same methods used in 2011 and 2014.

4.1 Coral and reef habitats

Benthic cover and composition - Point-intercept transects (PIT) were used to quantify benthic composition, recording the specific organisms or habitat types underlying each of 100 uniformly spaced points (50cm apart) along each transect. All corals and macroalgae were identified to genus. For survey points that did not intersect corals or macroalgae, the underlying habitat was be categorised as either, turf algae (< 5mm in height), crustose coralline algae (CCA), rubble or sand.

Topographic complexity - Topographic complexity was estimated visually at the start of each transect, using the six-point scale formalised by Wilson et al. (2007), where 0 = no vertical relief (essentially flat homogenous habitat), 1 = low and sparse relief, 2 = low but widespread relief, 3 = moderately complex, 4 = very complex with numerous fissures and caves, 5 = exceptionally complex with numerous caves and overhangs.

Page 14

Figure 2: Map showing the location of the study sites. (a) Geographic location of Elizabeth and Middleton reefs off Australia’s east coast. (b) Locations of the nine study sites on Middleton Reef. (c) Location of the seven existing study sites and three additional study sites at Elizabeth Reef. (d) Schematic diagram of reef cross-section showing the location of the three habitats and two depths sampled. 16 (out of 19) sites are explicitly matched to those surveyed in the 2011 (Pratchett et al. 2011) and 2014 (Hoey et al. 2014); Yellow circles = reef front locations, blue squares = back reef locations, and green stars = lagoon locations. “*” indicates new sites at Elizabeth Reef sampled in 2018.

Page 15

Table 2. Sites surveyed at Elizabeth and Middleton Reefs during the 2018 marine survey (24th February – 4th March, 2018). GPS co-ordinates are given for each site, as well as brief site descriptions

Reef Habitat Site Habitat description GPS

Elizabeth Reef 1 29º 57.161’S; Marked step down at 6m, then gentle Front 159º 01.295’E slope to >11m. Lots of deep holes and gutters at 4m.

Elizabeth Reef 2 29º 56.600’S; Gentle slope to 15m, then sandy Front 159º 01.100’E gutters. Deep holes and gutters at 4m.

Elizabeth Reef 3 29º 55.970’S; Distinct spurs up to 50m wide Front 159º 01.390’E separated by sand filled gutters. High relief at crest (3m) with wide gutters filled with coral boulders.

Elizabeth Reef 9 29o 57.423’S Gentle slope to 10m, then sharp drop- Front 159o 01.678’E off and lots of holes. Lots of deep holes and gutters at 4m

Elizabeth Reef 10 29o 57.366’S; Very wide slope with rolling hills and Front 159o 07.651’E wide valleys. Very flat at 3m with no distinct crest. Occasional deep gutters.

Elizabeth Back 4 29º 55.616’S; Moderate slope to 8m then steps down Reef 159º 02.406’E to sand at 10-12m. Deep gutters at 3- 4m, otherwise very flat.

Elizabeth Back 5 29º 55.113’S; Gentle slope to 8m, then steps down to Reef 159º 03.635’E sand at 11m. Lots of patch reefs out over sand. Deep holes and gutters at 4m.

Elizabeth Back 8 29o 54.613’S; Steep slope to sand/rubble at 12m. Reef 159o 04.132’E High relief at crest (2m) and lots of discontinuities in reef matrix.

Elizabeth Lagoon 6s 29º 56.148’S; Shallow lagoonal reef with very limited 159º 05.346’E relief, though mostly contiguous carbonate structure

Elizabeth Lagoon 6d 29º 56.079’S; Distinct rock/rubble bank that drops 159º 05.596’E down on to sand at ~7m, though deepens towards lagoon entrance

Elizabeth Lagoon 7 29º 56.170’S; Rubble reef flat with lots of arborescent 159º 03.097’E Acropora, dropping to 8m on sand.

Middleton Reef 5 29º 26.880’S; Located due north of the bow of Runic. Front 159º 03.310’E Gentle incline with lots debris.

Page 16

Middleton Reef 6 29º 28.637’S; Relatively narrow reef slope compared Front 159º 07.296’E to other reef front locations, and apparent drop off beyond deep transects.

Middleton Reef 7 29º 25.803’S; Very gentle incline (>100 m between Front 159º 07.988’E slope and crest habitat). Lots of debris from wreck in grooves near crest.

Middleton Reef 8 29º 29.086’S; Reef front location with characteristic Front 159º 04.597’E wide reef slope, and deep grooves extending to reef flat.

Middleton Reef 9 29º 27.262’S; Reef front location with characteristic Front 159º 08.532’E wide reef slope, and deep grooves extending to reef flat.

Middleton Back 1 29º 26.610’S; Located near wreck. Narrow reef slope Reef 159º 05.860’E which steps up sharply to reef crest.

Middleton Back 2 29º 27.177’S; Wide reef slope with gentle incline, but Reef 159º 05.061’E steps down on to sand at 12m.

Middleton Lagoon 3 29º 27.223’S; Discrete reef with extensive reef top 159º 03.940’E (1m depth) lactated among sand (3m depth).

Middleton Lagoon 4 29º2 6.574’S; Reef edge on northern margin of Blue 159º 06.901’E Holes, just behind two large bommies.

Coral recruitment – Densities of juvenile corals (≤5 cm maximum diameter, following Rylaarsdam 1983) were recorded within a 1m wide belt along the first 10m of continuous reef habitat on each transect. Densities of juvenile corals are intended to provide a proxy for population replenishment (Trapon et al. 2013a), incorporating variation in both settlement and early post-settlement mortality.

Coral health - To assess local impacts of the corallivorous crown-of-thorns starfish (Acanthaster cf. solaris), pin-cushion starfish (Culcita novaeguineae) and the corallivorous gastropod Drupella spp., as well as other large and potentially destructive coral predators such as excavating parrotfishes (Choat et al. 2006, Hoey and Bellwood 2008), any evidence of feeding scars were recorded 1m either side of the 50m transect path (i.e., 100m2). In addition to signs of corallivory, any

Page 17

evidence of adverse coral health, such as coral bleaching, and coral disease were also recorded within these 100m2 belt transects.

4.2 Non-coral invertebrates

Non-coral invertebrates, including potential coral predators (e.g., crown-of-thorns starfish, pin-cushion starfish, and coral snails) as well as ecologically and economically important mobile invertebrates (mainly, urchins and holothurians) were surveyed in a 2m wide belt along each transect, giving a sample area of 200m2. Coral predators are potentially important contributors to coral reef health and habitat structure, especially during periods of elevated densities of coral predation (Pratchett et al. 2014a). Importantly, outbreaks of crown-of-thorns starfish have occurred previously (in the 1980s) at Elizabeth and Middleton Reefs and caused extensive coral loss. The eastern Pacific crown-of-thorns starfish is now recognised as being morphologically and genetically distinct from the predominant Indian ocean species (Acanthaster planci) and is nominally referred to as Acanthaster cf. solaris (Pratchett et al. 2017). Acanthaster cf. solaris has been frequently sighted at Elizabeth and Middleton Reefs, as well as at Lord Howe Island and Ball’s Pyramid, though densities are generally very low.

Aside from coral predators, we also surveyed other conspicuous (non-cryptic) species of starfishes (asteroids), sea urchins (echinoids), sea cucumbers (holothurians), clams (Tridacna spp.) and sea anemones. For all these organisms, the number of individuals was counted within belt transects, providing density estimates to compare among surveys. Sea urchins can have a major influence on the habitat structure of coral reef environments (e.g., McClanahan and Shafir 1990; Eakin 1996). Like herbivorous fishes, larger species such as Diadema spp. may be important in removing algae that would otherwise inhibit coral growth and/or settlement (Edmunds and Carpenter 2001). At high densities, however, intensive grazing by sea urchins may have negative effects on habitat structure, causing significant loss of corals, reducing topographic complexity of reef habitats (McClanahan and Shafir 1990), can lead to a net erosion of the reef carbonates (Glynn et al. 1979; Eakin 1996). Sea urchins can at times be particularly abundant on high latitude reefs, and outbreaks of sea urchins have been recorded in this

Page 18

region. For example, populations of the collector urchin Tripneustes gratilla underwent an outbreak at Lord Howe Island in 2007-08, causing localized declines in the abundance of many red and brown algae (Valentine and Edgar 2010).

4.3 Coral reef fishes

Herbivorous reef fishes – Functionally important coral reef fishes (mostly, roving herbivores) were surveyed within a 5m wide path along replicate 50m transects (i.e., 250m2). All surgeonfishes (Acanthuridae), parrotfishes (Labridae), drummers (Kyphosidae), and rabbitfishes (Siganidae), were surveyed while deploying the transect tape. Individual fishes were identified to species and placed into 5cm size categories. Fish densities were converted to biomass using published length- weight relationships for each species, following Hoey and Bellwood (2009).

Endemic and site-attached reef fishes – Smaller more site-attached species from the families Chaetodontidae (butterflyfishes) and Pomacentridae (damselfishes), and juvenile double-header wrasse (Coris bulbifrons) were counted along a 2m wide path along each transect (i.e., 100m2) at every study site. Adult double-header wrasse (C. bulbifrons) were surveyed in a 5m wide belt along each transect (i.e., 250m2) at each study site. Mostly, we were interested in the local abundance of endemic species (Table 3).

Table 3: Endemic fishes that were surveyed at Elizabeth and Middleton Reef, following Hobbs and Feary (2007).

Species Common name Distribution

Coris bulbifrons Doubleheader Lord Howe, Norfolk, Middleton, wrasse Elizabeth, NSW

Chaetodon tricinctus Three striped Lord Howe, Norfolk, Middleton, butterflyfish Elizabeth

Amphiprion mccullochi McCulloch’s Lord Howe, Middleton, Elizabeth, NSW, anemonefish Norfolk

Epinephelus daemelii Black cod, Lord Howe, Middleton, Elizabeth, SE saddletail grouper Australia, Kermadec, Northern New Zealand

Page 19

Apex predators - To effectively survey larger and highly mobile reef-associated fishes (namely black cod and sharks) 250 - 500m long underwater visual transects were conducted along the reef at each study site, following Robbins et al (2006). Transects were run approximately parallel with the reef crest and covered a depth range of approximately 4 – 20m. All black cod (E. daemelii) and sharks (primarily the Galapagos shark, Carcharinus galapagensis) were recorded within 10-m of the transect path, giving a total sample area of 10,000m2 (1 hectare) for a 500m transect, and 5,000m2 (0.5 hectare) for a 250m transect. Two transects were conducted at each site, one 500m in length and the other 250m in length.

4.4 Data analyses

Recurrent sampling at the same locations and using the same sampling protocols to record a large number of variables (e.g., individual and combined cover of 85 categories of sessile organisms and substratum types, individual abundance of 37 genera of juvenile corals) is generating considerable and increasingly complex data. For the purpose of this report we have extracted only the most salient information necessary to assess current status and trends in a small number of key summary variables (e.g., live coral cover). However, very preliminary analyses were also conducted to assess the significance of temporal variation, relative to spatial variation within and among reefs, for key summary variables. These analyses were conducted using Generalised Linear Models (GLM) in R with increasing complexity of models that include year, reef, zone, and site or habitat. We then used model comparisons based on Akaike Information Criterion with sample correction (AICc) to select the best model and assessed the relative importance of the different predictor variables. Notably, these analyses show that year is an important variable (reflecting variation among surveys conducted in 2011, 2014 and 2018) in accounting for overall variation in coral cover, showing that the current sampling design is effective for assessing temporal trends in the ecosystem state and condition of these systems.

Page 20

5 Findings

5.1 Coral and reef habitats

Benthic cover and composition - In 2018, mean cover of hard (Scleractinian) corals recorded at Elizabeth Reef was 39.95% (±1.72 SE), ranging from 22.75 (±3.24 SE) within lagoon habitats, up to 46.00% (±2.31 SE) at pre-existing reef front locations (Figure 3a). Coral cover at Middleton Reef (26.62% ±1.87 SE) was lower than at Elizabeth Reef, mainly due to low coral cover (14.5% ±1.89 SE) in back reef locations (Figure 3a).

Overall coral cover at Elizabeth and Middleton Reefs has increased >40% since 2014 and is now higher (33.76% ±1.37 SE) than was recorded in 2011 (26.16% ±1.47 SE). The annual geometric rate of increase in coral cover from 2014 to 2018 equates to 9.2% for Elizabeth and 9.9% at Middleton Reef. Moreover, coral cover has increased in all habitats since 2014 (Figure 3a). The greatest increase has occurred within the lagoon at Middleton Reef, where recorded coral cover has more than doubled from 15.08% (±3.50 SE) in 2014 up to 32.42% (±3.72 SE) in 2018.

Mean cover of macroalgae at Elizabeth and Middleton Reefs was 11.43% (±1.56 SE) and generally similar (<10% cover) across most habitats (Figure 3b). Both the cover and the predominant type of macroalgae found at each site was largely unchanged since 2014.

Page 21

A 50 2011 45 2014 2018 40 35 30 25 20 15 Coral coverCoral (%) 10 5 0 Back Reef Lagoon Reef Front Back Reef Lagoon Reef Front Elizabeth Middleton

45 B 40 2011 2014 35 2018

10 Macroalgal coverMacroalgal (%) 5

0 Back Reef Lagoon Reef Front Back Reef Lagoon Reef Front

Elizabeth Middleton

Figure 3. Temporal changes (2011 - 2018) in average cover (±SE) of (a) scleractinian (hard) corals, and (b) macroalgae across three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs.

Page 22

Topographic complexity - Topographic complexity is generally low (consistently <3.0) across all sites at Elizabeth and Middleton Reefs. There was however evidence of very moderate increases in topographic complexity within some habitats (back reef and reef front) at Elizabeth Reef since 2014 (Figure 4). At Middleton Reef, however, structural complexity was largely unchanged relative to 2014, and was actually lower for Reef Front habitats in 2018 compared to 2014.

3 2011 2014 2.5 2018

1.5

0.5 Structural Complexity Score (out of 5)StructuralScoreComplexity of(out

0 Back Reef Lagoon Reef Front Back Reef Lagoon Reef Front Elizabeth Middleton

Figure 4. Temporal changes in the topographic complexity (±SE) of reef substratum across three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Topographic complexity was assessed visually using a 6-point scale (following Wilson et al. 2007). 0 = flat homogenous habitat, 5 = exceptionally complex.

Statistical analyses, using GLM and model selection, show that coral cover, macroalgal cover, topographic complexity, and densities of juvenile corals (which is discussed later) all vary with year of sampling (2011, 2014, and 2018), with Year

Page 23

included in the best models for all variables (Figure 5). Moreover, Year was most important variable for explaining variation in coral cover and topographic complexity. The most important variables for explaining variation in cover of macroalgae were Reef, Depth and Habitat, whereas variation in densities of juvenile corals were explained mostly by Site, and then Depth (or zonation).

Year Reef Depth Habitat Site

Live cover (%) of Scleractinian coral

Overall densities of juvenile corals

Live cover (%) of macroalgae

Topographic complexity

Figure 5. Variable selection and importance for alternative measures of coral habitat structure. All variables included in the selected model are coloured from yellow to red in terms of relative importance. Where there is no colour (i.e., Grey or white background colours) the factors was excluded from the best model.

Coral recruitment – The average density of juvenile corals in 2018 was 12.51 colonies per 10m2 (± 0.56 SE), which is slightly higher than was recorded in 2014 (11.26 ± 0.71 SE). However, temporal trends in the abundance of juvenile corals are variable among reefs and habitats (Figure 6). The highest densities of juvenile corals are generally recorded in reef front habitats at Elizabeth Reef, though the densities of juvenile corals in these habitats (14.79 ± 1.31 SE) were much lower in 2018 than had been recorded previously (Figure 6). At Middleton Reef, densities of juvenile corals in back reef habitats (13.37 ± 2.24 SE) were higher than have been recorded previously, but these increases were offset by a decline in the densities of juvenile corals on the Reef Front since 2014 (Figure 6).

Page 24

25 A 2011 2014 2018 20

5 Juvenile coralsJuvenile10sq.m) per(no. 0 Back Reef Lagoon Reef Front Back Reef Lagoon Reef Front Elizabeth Middleton

3.5 B 2011 2014 3 2018

2.5

1.5

1 Juvenile coralsJuvenile10sq.m) per(no. 0.5

Figure 6. Changes in (A) overall densities and (B) composition of juvenile corals (<5cm maximum diameter) three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Densities of juvenile corals (hard corals only) were recorded using replicate 10 x 1m belt transects.

Page 25

Juvenile coral assemblages at Elizabeth and Middleton Reefs are dominated by Acropora, Pocillopora, Isopora and Porites. The relative abundance of these taxa has been relatively consistent from 2011 to 2018, though there was an apparent decline in abundance of juvenile Acropora in 2014 (Figure 6). It should be noted that densities of juvenile Acropora recorded in 2018 (2.24 ± 0.07 SE) were equivalent to those recorded in 2011 (2.39 ± 0.08 SE). Also, densities of juvenile Porites increased from 2011 (0.75 ± 0.13 SE) to 2014 (1.94 ± 0.33 SE), and remained high in 2018 (Figure 6). Local variation in the abundance of Isopora is much more pronounced than for other dominant taxa (as evident by the very large SE), reflecting patchiness in their distribution and abundance, but also their disproportionate abundance in specific habitats, shallow reef fronts. Overall densities of juveniles for other less common genera (e.g., Seriatopora, Stylophora and Acanthastrea) have remained fairly constant across the three survey periods (2011, 2014 and 2018).

Coral health - There were no definitive or confirmed incidences of coral disease recorded at Elizabeth and Middleton Reefs in 2018, though recent and extensive mortality of several distinct colonies of Acropora and Isopora corals in the lagoon (Site 3) at Middleton Reef, may have resulted from disease (most likely White Syndrome). The problem with white syndrome is that rapid tissue loss may occur in absence of any clear signs of disease (Sussman et a. 2008, Sweet and Bythell 2012) and are indistinguishable from tissue removal caused by coral predators when observed only after whole coral mortality. We could not clearly establish the cause of this recent coral mortality, but given the apparent lack of coral predators (e.g., crown-of-thorns starfish) at this site, the most likely causes are coral disease and/ or coral bleaching (see below).

Bleached corals were prevalent and conspicuous, especially in the back reef (Site 3) and lagoon (Site 7) at Elizabeth Reef, when surveyed on February 19th, 2018. Bleaching was severe, but restricted to select coral species, mainly Stylophora cf. subseriata (Figure 7), which is among those corals that are known to be particularly prone to bleaching (McClanahan et al. 2004). This is concerning given that it there was significant potential for further accumulation of heat stress prior to seasonal

Page 26

cooling of ocean temperature in April/May. Accordingly, we compiled NOAA satellite sea-surface temperature (SST) data to explicitly consider the accumulated heat stress (measured as degree heating weeks; Liu et al. 2003) in 2018 compared to other recent years. We find that heat stress continued to accumulate well beyond the timing of our surveys (late February/ early March), when bleaching was already apparent (Figure 7), and the risk of bleaching is higher in 2018 than it has been in recent years.

Figure 7. Bleaching risk for Elizabeth Reef; Severe, but selective, coral bleaching was observed within the lagoon (Site 7) at Elizabeth Reef on February 19th. 2018. Virtually all colonies of Stylophora (shown here) were completely bleached at this location, along with a proportion of colonies of Montipora and Acropora. Accumulated heat stress (in Degree Heating Weeks) for Elizabeth Reef in 2018 (DHW>3) is the higher than recorded in previous years (2016-17).

Page 27

5.4 Non-coral invertebrates

Coral predators - Densities of coral-feeding invertebrates, specifically crown-of- thorns starfish (Acanthaster cf. solaris), the pin cushion starfish (Culcita novaeguineae) and coral snails (Drupella spp.) were very negligible at Elizabeth and Middleton Reefs. In 2018, only one crown-of-thorn starfish, one pin cushion starfish, and two individual coral snails were recorded across all 155 transects. Despite increased search area (owing to the addition of three new sites at Elizabeth Reef), these numbers are lower than were recorded in 2011 and 2014, though abundances of corallivorous species have been very low throughout (Table 4). The single crown-of-thorns starfish that was recorded in 2018 (as well as one additional individual that was found just off the transect) was located in the back reef (Site 4) at Elizabeth Reef. Notably, this is the same site where crown-of-thorns starfish have been sighted during previous surveys. This suggests that there is a small persistent remnant population restricted to this area that are being re- surveyed each year.

In accordance with the low densities of A. cf. solaris, Drupella spp. and C. novaeguineae, the incidence of feeding scars on live coral colonies across all 155 transects was negligible. Recent coral injuries that were detected, were not consistent with feeding scars caused by the above mentioned coral-feeding invertebrates, but represent physical removal of many (if not all) branches, presumably caused by feeding activities (i.e., ‘cropping’) by fishes, most likely caused by the excavating parrotfish, Chlorurus microrhinos and C. frontalis (Hoey et al. 2014). The overall incidence of this cropping in 2018 (29 colonies, 0.23 colonies per transect) was substantially lower than recorded in 2014 (100 colonies, 0.81 colonies per transect), and largely restricted to exposed reef front locations at Middleton Reef. Moreover, affected corals were almost exclusively Acropora. This damage is not cause for concern and is likely to reflect the normal background levels of coral predation by excavating parrotfishes.

Page 28

Table 4: Total abundance of coral-feeding invertebrates recorded at Elizabeth and Middleton Reefs, from 2011-2018. Species 2011 2014 2018 Total

No. transects 124 124 155 403

Crown-of-thorns starfish 1 4 1 6 (Acanthaster cf. solaris)

Pin cushion starfish 3 6 1 10 (Culcita novaeguineae)

Coral snails 5 3 2 10 (Drupella spp.)

Sea cucumbers (Holothurians) - A total of 839 sea cucumbers were recorded across all transects (n = 155) in 2018, corresponding with a mean density of 5.41 (± 0.95 SE) individuals per 200m2, which is higher than recorded in 2011 (3.29 ± 0.77 SE) and 2014 (5.06 ± 0.93 SE). The dominant species were Holothuria edulis (437/ 839 individuals) and Holothuria atra (319/ 839), both of which exhibit increasing trends in abundance (Figure 8). Overall densities of sea cucumbers were highest in the lagoon, coinciding with the availability of sand, as opposed to consolidated reef matrix. Densities of sea cucumbers were also higher at Middleton Reef than at Elizabeth Reef (Figure 8).

Sea urchins (Echinoids) - >10,000 sea urchins (mostly Echinostrephus sp. and Diadema savignyi) were counted across 155 transects at Elizabeth and Middleton Reefs in 2018. Diadema savignyi (which was previously misidentified as D. setosum) is a large, mobile and nocturnally active herbivore that is most abundant in the Back Reef at Middleton Reef (Figure 10a). Densities of D. savignyi recorded in 2018 within this habitat (17.5 ± 6.63 SE urchins per 200m2) are a fraction of maximum densities (171.0 ± 47.7 SE) recorded in 2014. Densities of Echinostrephus sp., which is a small burrowing urchin with long thin spines are found predominantly outside of the lagoon, and are common in both Back Reef and Reef Front habitats (Figure 9). Overall densities of Echinostrephus sp. were higher in 2018 than recorded previously, largely due to high densities in the Back Reef at Middleton Reef (Figure 10).

Page 29

35 A 2011 2014 30 2018 25

10 Density200sq.m) per(no. 5

0 Back reef Lagoon Reef front Back reef Lagoon Reef front Elizabeth Middleton

4 B 3.5 3 2.5 2 1.5 1

Density200sq.m) per(no. 0.5 0

Figure 8. Changes in (A) overall densities and (B) composition of sea cucumbers at three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Densities of sea cucumbers are recorded along replicate 50 x 2m belt transects.

Page 30

250 2011 200 2014 2018 150

100

50 Density200sq.m) per(no.

0 Back reef Lagoon Reef front Back reef Lagoon Reef front Elizabeth Middleton

Figure 9. Changes in densities of Diadema savignyi (pictured above) within three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Densities of urchins are recorded along replicate 50 x 2m belt transects.

Page 31

180 160 140 120 100 80 60 40 20 Density200sq.m) per(no. 0 Back reef Lagoon Reef front Back reef Lagoon Reef front

Elizabeth Middleton

Figure 10. Changes in densities of Echinostrephus spp. within three different habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Densities of urchins are recorded along replicate 50 x 2m belt transects.

5.3 Coral reef fishes

Roving herbivores - The overall abundance of herbivorous fishes was very similar between reefs in 2018, averaging 58.8 (± 6.1 SE) fishes/250m2 on Elizabeth and 60.0 (± 5.3 SE) fishes/250m2 on Middleton Reef (Figure 11a). The abundance of roving herbivorous fishes did differ between habitats and depths (Figure 11a), being consistently higher on the reef crest than the reef slope at exposed reef front sites on both Elizabeth and Middleton Reefs, but displayed limited difference among depths at either the lagoon or back reef (Figure 11a). These patterns are consistent with numerous studies on tropical and subtropical reefs that have demonstrated the abundance and biomass of herbivorous fishes is generally greatest on shallow habitats on exposed reef fronts and decreases with both depth and decreasing wave exposure (e.g., Russ 1984, 2003; Hoey and Bellwood 2008; Hoey et al. 2013, 2018).

In contrast to abundance, the biomass of herbivorous fishes differed between reefs, averaging 17.4 (± 2.0 SE) kg/250m2 on Elizabeth and 24.4 (± 3.8 SE) kg/250m2 on Middleton Reef (Figure 11b). In addition to these differences in biomass between the two reefs, herbivore biomass differed among habitats and

Page 32

reef zones, with biomass on the reef crest within the exposed reef front and back reef habitats being 2- to 5-fold greater than the biomass of the shallow (i.e., reef crest) habitat within the lagoons at both reefs (Figure 11b). Herbivore biomass was more consistent among habitats on the reef slope. The different patterns in abundance and biomass reflect differences in the body size among habitats. This distinction is important as herbivore biomass has been shown to be a better predictor of ecological impact than abundance (e.g., Bonaldo and Bellwood 2008; Lokrantz et al. 2008; Hoey and Bellwood 2009). In contrast, comparable abundances but lower biomass of herbivorous fishes within the lagoonal habitats is reflective of the generally smaller bodied fishes, notably juvenile fishes, and highlights the importance of the lagoons on these reefs as nursery habitats.

Page 33

Figure 11. Spatial variation in (a) the abundance, and (b) the biomass of roving herbivorous fishes among reefs (Elizabeth and Middleton), habitat zones (Back reef, lagoon, reef front), and depths (crest, slope) in February-March 2018 (see Figure 2 and Table 2 for details). Data are means ± SE. The dashed lines represent overall reef means.

Page 34

Overall, the abundance of herbivorous fishes recorded at Elizabeth and Middleton Reefs in 2018 were remarkably similar to those recorded in both 2011 and 2014 (Figure 11a; Elizabeth: 2011: 67.3 fishes/250m2, 2014: 54.8 fishes/250m2, 2018: 58.8 fishes/250m2; Middleton: 2011: 72.8 fishes/250m2, 2014: 59.6 fishes/250m2, 2018: 60.0 fishes/250m2). Similarly, the abundance of herbivores within the back reef and exposed reef front sites were relatively stable among surveys (2011, 2014 and 2018), whereas abundances of herbivorous fishes in the lagoon sites declined from 2011 to 2014, and then increased again in 2018 (Figure 12a). The observed changes in the lagoon sites was primarily attributable to changes in the abundance of small-bodied grazing fishes, primarily juvenile parrotfishes, in these habitats (Figure 12b), and may reflect temporal variation in the recruitment of these species. Interestingly, previous declines in the abundance of grazing fishes at both Elizabeth and Middleton Reefs from 2011 to 2014, have largely been reversed with our abundance estimates for 2018 being broadly similar to those recorded in 2011 (Figure 12b). There have been some declines in the abundance of browsing fishes at both Elizabeth and Middleton Reefs since 2014, however the abundances of browsing fishes in 2018 are as high or higher than those from 2011 (Figure 12c).

The biomass of herbivorous fishes was relatively consistent at Middleton Reef over the three surveys (2011: 31.4 kg/250m2; 2014: 25.2 kg/250m2; 2018: 24.5 kg/250m2; Figure 13a). In contrast, the biomass of herbivorous fishes at Elizabeth reef declined by 43%, from 30.7 kg/250m2 in 2014 to 17.5 kg/250m2 in 2018 (Figure 12a). Despite this decline the biomass of herbivorous fishes on Elizabeth Reef in 2018 was still greater than the herbivore biomass recorded in 2011 (13.2 kg/250m2). This temporal variation in total herbivore biomass was almost solely attributable to changes in the biomass of browsing fishes (primarily the Pacific Chub Kyphosus sectatris (formerly Kyphosus pacificus) and the Spotted Sawtail Prionurus maculatus) (Figure 13c). In contrast, there was limited temporal variation in the biomass of grazing fishes between 2014 and 2018 (Figure 13b).

Page 35

Figure 12. Temporal variation (2011 - 2018) in the mean abundance of roving herbivorous fishes across three habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. (a) Total herbivorous fishes, (b) Grazing fishes, (c) Browsing fishes. Error bars represent 1 SE.

Page 36

Figure 13. Temporal variation (2011-2018) in the mean biomass of roving herbivorous fishes across three habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. (a) Total herbivorous fish biomass, (b) Grazing fish biomass, (c) Browsing fish biomass. Error bars represent 1 SE.

Page 37

A principal component analysis (PCA) revealed strong spatial structuring, but limited temporal change, in the taxonomic composition of the herbivorous fish assemblages on Elizabeth and Middleton reefs (Figure 14). The first two axes of the PCA explained 39.4 % of the total variation in composition of herbivore assemblages, with the exposed reef (or reef front) crest sites clearly separated from all lagoonal sites along the first principal component. The reef front crest sites were characterised by a high biomass of the macroalgal browsing Kyphosus spp, large excavating parrotfishes (Chlorurus frontalis, C. microrhinos) and to a lesser extent the browsing Prionurus maculatus and scraping parrotfish Scarus altipinnis. In contrast, the lagoonal sites were characterised by the smaller excavating parrotfish (Chlorurus spilurus, formerly Chlorurus sordidus), and a diversity of smaller scraping parrotfishes (Scarus chameleon, S. globiceps, S. psitticus; Figure 14).

Interestingly, there was no relationship between the biomass of roving herbivorous fishes and either the cover of macroalgae, the cover of live scleractinian (hard) coral, or the cover of bare space (i.e., turf algae and crustose coralline algae) (Figure 15). While numerous studies have linked the abundance and/or biomass of herbivorous fishes to the cover of live coral and/or the structural complexity of the habitat, and others have reported negative correlations between the cover of macroalgae and the biomass of herbivorous fishes across a range of spatial scales (e.g., Williams et al. 2001; Wismer et al. 2009), we found no evidence for such relationships at Elizabeth and Middleton Reefs. This is largely consistent with previous surveys at these reefs (Hoey et al. 2014), and may reflect the limited variation in the cover of these benthic variables, or that other factors (such as wave action and larval supply) that may be influencing these relationships. The lack of relationship with either macroalgae or live coral while unexpected, does not mean that herbivorous fishes are unimportant in the ecology and resilience of coral reefs at Elizabeth and Middleton Reefs. Clearly, further research is required to understand the ecological functioning of these high latitude reefs, especially what are the determinants of macroalgal abundance and biomass (Choat et al 2006, Pratchett et al. 2011, Hoey et al. 2011, 2018).

Page 38

Figure 14. Principal component analysis (PCA) showing variation in the composition of roving herbivorous fish assemblages between years, habitats and depths. Symbols indicate sampling year; triangles - 2011, circles – 2014, squares - 2018. Colours indicate habitat; dark green – lagoon slope, light green – lagoon crest, dark blue – back reef slope, light blue – back reef crest, bright yellow – exposed reef crest, dull yellow – exposed reef slope. Analysis was based on the covariance matrix of the mean biomass (log- transformed) of each species at each site.

Page 39

Figure 15. Relationships between the herbivore biomass and the cover of (a) macroalgae, (b) live coral, and (c) bare space (the sum of algal turfs and CCA). Each point represents the mean of 4 transects at each depth at each of 19 sites. triangles: Elizabeth Reef, circles: Middleton Reef. Colours represent the three habitats shown in Figure 2. Yellow: exposed reef front; Blue: back reef; Green: lagoon.

Page 40

Endemic and site-associated fishes - The endemic anemone fish Amphiprion mccullochi is a relatively minor component of the total pomacentrid community on Elizabeth and Middleton Reefs, with an average density of 0.31 fish/100m2 on Elizabeth Reef and 0.03 fish/100m2 on Middleton Reef (Figure 16). In contrast, the endemic three-striped butterflyfish Chaetodon tricinctus was the most abundant of 22 butterflyfish species recorded on these reefs with a mean density of 1.28 fish/100m2 on Elizabeth Reef and 1.15 fish/100m2 on Middleton Reef (Figure 16).

A. mccullochii is a habitat specialist, being only found in close association with the anemone Entacmaea quadricolor. The distribution of the A. mccullochi is, therefore, restricted by the availability of its host anemone. At Elizabeth and Middleton Reefs this anemone is restricted primarily to lagoonal areas and is reflected in the distribution on A. mccullochi (Figure 17a). The low and variable abundance of A. mccullochi precluded any meaningful analyses. In contrast, the three-striped butterflyfish C. tricinctus was more widely distributed among habitats, being recorded across all habitats on both reefs (Figure 17b). Although there is some debate over the diet of C. tricinctus (Pratchett et al. 2014b), the abundance of this species was positively related to live coral cover suggesting it relies on live coral for food and/shelter. Like many other butterflyfishes, such reliance on live coral likely make populations of this species will be susceptible to changes in live coral cover.

The overall abundances of both A. mccullochi and C. tricinctus have remained relatively stable from 2011 to 2018 (Figure 18). The abundance of A. mccullochi was relatively stable on Elizabeth Reef, but showed a sharp decline on Middleton Reef from 2014 to 2018 (Figure 18a). This decrease is no cause for alarm as the abundances are now back to 2011 levels, and several individuals were seen outside the transect boundaries in the lagoon sites at Middleton reef. The densities of C. tricinctus were relatively consistent from 2014 to 2018 on both Elizabeth and Middleton Reefs, and remained above the 2011 abundances (Figure 18b). There was a relatively small decline in abundances of C. tricinctus on Elizabeth Reef from 1.89 ind/100m2 in 2014 to 1.28 ind/100m2 in 2018, that was driven by a decline in abundances on exposed reef front sites (Figure 18b), despite increases in live coral cover over the same period (Figure 3).

Page 41

Figure 16. Relative abundances of members of the (a) Pomacentridae (damselfishes), and (b) Chaetodontidae (butterflyfishes) recorded across all transects on Elizabeth and Middleton Reefs in February/March 2018. Arrows indicate the relative abundances of the endemic species Amphiprion mccullochi and Chaetodon tricinctus. Data are means ± SE.

Page 42

Figure 17. Spatial variation in the abundance of (a) Amphiprion mccullochi, and (b) Chaetodon tricinctus among reefs (Elizabeth and Middleton), habitat zones (back reef, lagoon, reef front), and depths (crest, slope) in February/March 2018 (see Figure 2 and Table 2 for details). Data are means ± SE.

Page 43

Figure 18. Temporal variation (2011 - 2018) in the mean abundance of (a) Amphiprion mccullochi, and (b) Chaetodon tricinctus across three habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Error bars represent 1 SE.

Page 44

The endemic double-header wrasse Coris bulbifrons was relatively abundant across both Elizabeth and Middleton Reefs in 2018, with 206 adults and 39 juveniles (<10cm Total Length) being recorded across all transects, compared to 232 adults and 14 juveniles in 2014. Juvenile C. bulbifrons were largely restricted to the back reef and lagoon habitats (Figure 18a), while adult C. bulbifrons were recorded across all habitats and depths at both reefs (Figure9).

The mean densities of both juvenile and adult C. bulbifrons were relatively stable at Elizabeth and Middleton reefs over the past 4-years. The overall densities of juvenile C. bulbifons changed little at Elizabeth Reef from 2014 to 2018, although there was a decline in abundance from 0.09 ind/100m2 in 2014 to 0.03 ind/100m2 to 2018 on Middleton Reef (Figure 20a). This decline likely represents variation in recruitment success among years. The densities of adult C. bulbifons remained unchanged on Middleton Reef from 2014 to 2018, but showed a small decline on Elizabeth Reef over the same period (2014: 0.78 ind/100m2; 2018: 0.44 ind/100m2; Figure 20b). This decrease in adult C. bulbifrons was solely attributable to a decline in abundances within the lagoon sites.

Page 45

Figure 19. Spatial variation in the abundance of (a) juvenile, and (b) adult Coris bulbifrons among reefs (Elizabeth and Middleton), habitat zones (back reef, lagoon, reef front), and depths (crest, slope) in February-March 2018 (see Figure 2 and Table 2 for details). Data are means ± SE.

Page 46

Figure 20. Temporal variation (2011 - 2018) in the mean abundance of (a) juvenile, and (b) adult Coris bulbifrons across three habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Error bars represent 1 SE.

Page 47

Apex predators - High densities of Galapagos reef sharks (Carcharhinus galapagensis) were again recorded at Elizabeth and Middleton Reefs in 2018. The average abundance of C. galapagensis across all habitats was 10.2 individuals per hectare (± 4.0 SE) on Elizabeth Reef and 8.9 individuals/ha (± 2.4 SE) on Middleton Reef (Figure 21a). These abundances represent an increase at Elizabeth Reef (2014: 6.58 individuals/ha ± 1.69 SE) and a decrease at Middleton Reef (2014: 17.11 individuals/ha ± 2.99 SE). The most pronounced increases were recorded within the lagoon sites on Elizabeth Reef, while moderate declines were recorded in the back reef sites on both reefs and the reef front on Middleton reef (Figure 21a). The vast majority of C. galapagensis recorded were relatively small (< 120cm; Figure 21) with the largest individual recorded being 200cm total length. These size distributions largely reflected those from 2014 (Figure 22). Given that C. galapagensis reaches up to 300 cm in length, does not mature until 210-250 cm, and is born at 60-80 cm (Last and Stevens 1994), these data indicate that the large populations at Elizabeth and Middleton Reefs comprise mostly immature individuals, and that these reefs are likely to represent an important nursery area for this species.

Page 48

Figure 21. Temporal variation (2011 - 2018) in the mean abundance of (a) the Galapagos Reef Shark and (b) the Black Cod across three habitat types (see Figure 2 and Table 2 for details) at Elizabeth and Middleton Reefs. Error bars represent 1 SE.

Page 49

Figure 22. Comparison of the size structure of Galapagos Shark (Carcharhinus galapagensis) populations between Elizabeth and Middleton Reefs in (A) 2014, and (B) 2018. Size at birth is 60-80cm, size at maturity is 210-230cm (males) and 250cm (females).

Page 50

Black cod (Epinephelus daemelii) were recorded across most sites at Elizabeth and Middleton Reefs (Figure 21b). The average abundance of E. daemelii across all habitats in 2018 was 1.50 individuals per hectare (± 0.35 SE) at Elizabeth Reef, and 1.41 individuals per hectare (± 0.55 SE) at Middleton Reef. These represent declines of approximately 40% from 2014 levels (Elizabeth: 2.50 (± 0.44 SE) individuals per hectare; Middleton: 2.24 (± 0.53 SE) individuals per hectare). Concerningly, theses declines were were widespread with abundances of E. daemelii decreasing in all three habitat zones on both reefs, with no E. daemelii being recorded at the lagoon sites on Middleton Reef in 2018. The reasons for these declines are not apparent but clearly warrant further investigation.

Page 51

6 Conclusions

Elizabeth and Middleton Reefs are important examples of sub-tropical reefs, supporting unique assemblages of both tropical and temperate species (Hobbs et al. 2008, Hoey et al. 2011). While not exposed to the frequency or intensity of disturbances affecting tropical reefs, subtropical reefs are extremely vulnerable to acute disturbances, such as coral bleaching and disease, storms, and crown-of- thorns starfish outbreaks (e.g., Harriott 1995; Riegl 2003; Harrison et al. 2011) because their capacity for recovery and resilience is likely to be limited both by isolation (e.g., Gilmour et al. 2013) and environmental constraints on population dynamics (Anderson et al. 2015). Current rates of increasing coral cover at Elizabeth and Middleton Reefs (9-10% per year) are comparable to rates of coral recovery recorded on the Great Barrier Reef (e.g., Lourey et al. 2000), though it is clear that coral recovery since outbreaks of crown-of-thorns starfish in this region in the 1980s has been extremely protracted.

6.1 Coral and reef habitats

Coral cover at Elizabeth and Middleton Reef is currently high (33.76% ±1.37 SE), especially compared to highly depressed mean coral cover across the Great Barrier Reef in the aftermath of the recent back-to-back coral bleaching events (Hughes et al. 2017a, 2017b, 2018). Even more importantly, coral cover at Elizabeth and Middleton Reefs is increasing, whereas coral cover is generally declining across most major coral reef regions (e.g., Bellwood et al. 2004; Bruno and Selig 2007), and declined further due to recent, unprecedented, severe and recurrent bleaching at many locations (Hughes et al. 2018).

Densities of juvenile corals at Elizabeth and Middleton Reefs are very low compared to low latitude reef systems, but similar to those from Lord Howe Island. These low rates of replenishment appear to be characteristic of high latitude reefs (Harriott 1992; Hoey et al. 2011; Bauman et al. 2014), and coupled with the lower coral growth rates on subtropical reefs (Harriott 1999; Anderson et al. 2012) and isolation (Gilmour et al. 2013) are likely to limit the recovery potential of these reefs following disturbance. Collectively these factors are likely to have contributed to the

Page 52

protracted recovery of coral communities following the crown-of-thorns starfish outbreak in the late 1980’s and early 1990’s.

Given the high and increasing coral cover at Elizabeth and Middleton Reefs, it appears that these locations were largely unaffected by recent spikes in global ocean temperatures that caused wide-spread and severe coral bleaching (Hughes et al. 2017b). However, severe bleaching of select corals was conspicuous in some locations during recent surveys. Moreover, the temperature stress that occurred in 2018 (>3 DHW) was higher than recorded at these locations in 2016 and 2017. For context, there is a 40% probability of severe mass bleaching on the GBR at 3DHW (Hughes et al. 2017a). This might provide significant imperative to re-visit these locations and establish vulnerability to temperature stress and coral bleaching. These locations are particularly vulnerable to coral bleaching or any other acute disturbances, given low rates of population replenishment, which will greatly constrain recovery.

6.2 Non-coral invertebrates

Coral predators – While there have been previous outbreaks of crown-of-thorns starfish at Elizabeth and Middleton Reefs (Australian Museum 1992; Harriot 1998), which devastated local coral assemblages with very protracted recovery, current densities of A. cf. solaris (as well as other invertebrate corallivores) are very low. There is, however, a persistent low-density population of A. cf. solaris at Elizabeth Reef. Given that outbreaks of crown-of-thorns starfish may arise through the gradual accumulation of individuals over several successive cohorts (Pratchett 2005), culling these low-density populations may greatly reduce the risk of future outbreaks (Pratchett et al. 2014a). Conversely, there is little to no evidence that low-density populations of crown-of-thorns starfish have beneficial effects on the structure or diversity of coral assemblages (Pratchett 2010).

Sea cucumbers - The management and isolation of Elizabeth and Middleton Reefs appears to be having positive effects on sea cucumber populations, which are overexploited in most other locations (Purcell 2014, Purcell et al. 2014). Densities of H. whitmaei (listed as endangered on IUCN Redlist) recorded in 2018

Page 53

were lower than recorded in 2014 (Figure 9), but it is unknown whether this is reflective of reef-wide population declines and other predominant species were more abundant in 2018 than recorded in 2011 or 2014. It is also important to realise that these surveys are designed to assess fish and benthic communities within consolidated reef, or coral-dominated, habitats, whereas accurate assessment of sea cucumber populations would require dedicated sampling of shallow reef flat areas and sand-dominated habitats.

Urchins - Although they are considered functionally important in some locations (e.g., Diadema on Caribbean reefs), high densities of urchins on Indo-Pacific reefs are often viewed as evidence of reef degradation (Bellwood et al. 2004). Densities of D. savignyi within the back reef at Middleton reef have actually declined since 2014, though there have been corresponding increases in the abundance of Echinostrephus spp. within this habitat. Notably, these habitats are already characterized by high levels of bioerosion, potentially attributable to sustained high densities of Diadema, and Echinostrephus spp. may further exacerbate erosion and destabilise reef structures. Echinoids, and echinoderms in general, are renowned for their marked population fluctuations (Uthicke et al. 2009), but the ecological importance and consequence of these population fluctuations does warrant increased attention.

6.2 Reef fishes

Fishes on coral reefs are being increasingly viewed in terms of their ecological roles, or functions, irrespective of their taxonomic affinities (e.g., Richardson et al. 2018). This shift in focus has been driven by the realisation that the removal of key functional groups of fishes through fishing activities has underpinned the degradation of coral reef systems worldwide (e.g., Hughes 1994; Rasher et al. 2013). Reductions in predatory fishes have been suggested to cause meso- predator release and potentially drive trophic cascades, although evidence for such effects is limited (e.g., Casey et al. 2017). Reductions in predators have, however, been related to increased incidence of outbreaks of crown-of-thorns starfish in the tropical Pacific, and the subsequent loss of live coral cover (Sweatman 2008; Dulvy et al. 2004), as well as changes in the foraging behaviour of herbivorous fishes

Page 54

(Rizzari et al. 2014, Rasher et al. 2017). Perhaps the most pervasive effect of fishing on the functioning of coral reefs is the reduction in herbivorous fish populations that have underpinned shifts from coral- to macroalgal-dominated reefs in many regions (Hughes 1994; Rasher et al. 2013; Graham et al. 2015). Coral reef fish communities appear to be relatively stable at Elizabeth and Middleton Reefs with similar abundances recorded from 2011 to 2018 (Pratchett et al. 2011; Hoey et al. 2014). The only exception to this was the black cod (E. daemelii) which exhibited small but consistent declines across all habitats at both Elizabeth and Middleton Reefs from 2014 to 2018.

The Black cod (E. daemelii) has declined throughout much of its geographic range, largely due to overfishing. At Elizabeth and Middleton Reefs, the overall abundance of black cod (across all sites) averaged 1.50 and 1.44 individuals per hectare respectively, and there has been a small but consistent decline in densities recorded at the same sites in 2011 and 2014 (Pratchett et al. 2011; Hoey et al. 2014). Although the current densities of Black cod are within the range of densities recorded by Oxley et al. (2004) and Choat et al (2006), the consistency of the declines across all habitats is cause for concern. It appears unlikely that these declines are a direct result of fishing activities as other species that are likely to be affected by line fishing (e.g. Galapagos shark, double-header wrasse) did not decline in abundance. The smallest Black cod observed during the present surveys was 40cm in length, suggesting that recruitment to the population may be limited, or that recruitment habitats were not adequately captured by our surveys (e.g., they may recruit to deeper water). Continued and extensive long-term monitoring is critical to assess the fate of Black cod at Elizabeth and Middleton Reefs, considered one of the last strongholds for this species. Importantly, more intensive research is also required to assess the natural dynamics of local populations (e.g., recruitment rates, recruitment/nursery habitats, growth and longevity), which will be critical for assessing the overall vulnerability of these populations. Any future declines, no matter how slight, could threaten the sustainability of this population.

The structure and resilience of coral reef ecosystems are fundamentally dependent on the actions of herbivorous fishes. Through their grazing activities, herbivorous fishes have been estimated to remove up to 90 percent of the net algal production

Page 55

from shallow coral reefs (Polunin and Klumpp 1992; Ferreira et al. 1998). In doing so herbivorous fishes inhibit the proliferation of algal biomass (Lewis 1986), minimize algal-coral competition (Hughes et al. 2007; Rasher and Hay 2010) and facilitate coral settlement through the provision of suitable settlement substrata (Hughes et al. 2007; Hoey et al. 2011). On reefs with relatively intact herbivorous fish assemblages, populations of herbivores have been shown to increase in response to large-scale coral loss and consequent increases in algal cover (Adam et al. 2011, Gilmour et al. 2013) thereby compensating the increased algal production and promoting the recovery of coral assemblages following disturbance (Bellwood et al. 2004; Mumby and Steneck 2008).

The herbivorous fish assemblages of Elizabeth and Middleton Reefs represent a unique mix of both tropical and temperate species, and appear to have changed little over the past 7 years (Pratchett et al. 2011; Hoey et al. 2014). Collectively, the densities of all herbivorous fishes on Elizabeth and Middleton Reefs were broadly comparable to lower latitude reefs of the central and northern Great Barrier Reef (Wismer et al. 2009) and substantially greater than those of Lord Howe Island, 260 km to the south (e.g., Hoey et al. 2011). There were, however, marked differences in the functional composition of herbivorous fish assemblages between these different reef systems. Herbivorous fish assemblages on Elizabeth and Middleton Reefs were dominated by macroalgal browsing species, in particular Kyphosus sectatris. On lower latitude, or tropical, reefs macroalgal browsing species typically account for < 10% of the herbivorous fish biomass (e.g., Hoey and Bellwood 2010a, b). While these fishes have been suggested to play a critical role in removing macroalgal biomass and potentially reversing phase-shifts on low latitude reefs (Hoey and Bellwood 2011; Rasher et al. 2013), their role limiting the cover and biomass of macroalgae on higher latitude reefs, such as Elizabeth and Middleton Reefs, is largely unknown. For example, the two browsing species Kyphosus sectatris and Prionurus maculatus accounted for >70 % of the total herbivorous fish biomass on Elizabeth and Middleton reefs, but we currently know very little of their diet, feeding behaviour, functional impact. The role these fishes play in limiting the abundance and biomass of macroalgae on high latitude reefs is currently unknown and is an area that requires specific targeted research if we are to understand the process that structure these unique reef systems.

Page 56

Together with the high biomass of browsing fishes, the biomass of grazing fishes on Elizabeth and Middleton Reefs has been stable over the past 7 years at a level that is approximately 25% of that on low latitude reefs. Grazing fishes, and in particular scraping and excavating parrotfishes, perform a key role in clearing space for the settlement of benthic organisms such as corals (Mumby 2006; Hoey and Bellwood 2008, Trapon et al. 2013a,b). Although there has been some debate regarding the nutritional target of parrotfishes, a recent synthesis indicates they are targeting protein-rich epiphytic and endolithic phototrophs (primarily cyanobacteria, Clements et al. 2017). Irrespective of their nutritional target, intensive feeding by parrotfishes and other grazing fishes is critical to maintaining algal communities in a cropped state (e.g., Bellwood et al. 2006, Hughes et al. 2007). Interestingly, we found no relationship between the biomass herbivorous fishes and the cover of macroalgae, live coral, or bare space (i.e., algal turfs and crustose coralline algae). Focused and intensive research will be required to disentangle the complexities of the relationships between herbivorous fishes and the condition and composition of benthic assemblages on these unique reefs.

Page 57

7 Acknowledgements

The latest surveys (February-March 2018) at Elizabeth and Middleton Reefs were conducted using M.V. Iron Joy, alongside parallel surveys undertaken by Reef Life Survey. We are grateful to the entire crew, especially skipper Rob Benn.

Field surveys were commissioned by the Director of National Parks, with additional in-kind support from the Australian Centre of Excellence for Coral Reef Studies. We also gratefully acknowledge the guidance and contributions of Cath Samson, who would have been a very valuable person to accompany us during this expedition.

Page 58

8 References

Adam, T.C., Schmitt, R.J., Holbrook, S.J., Brooks, A.J., Edmunds, P.J., Carpenter, R.C. and Bernardi, G. (2011). Herbivory, connectivity, and ecosystem resilience: response of a coral reef to a large-scale perturbation. PloS one, 6, e23717.

Australian Museum. 1992. Reef Biology - A survey of Elizabeth and Middleton Reefs, South Pacific. Kowari 3. The Australian Museum, Sydney. An Australian National Parks and Wildlife Service publication.

Anderson, K., Pratchett, M., and Baird, A. (2012). Summer growth rates of corals at Lord Howe Island, Australia. Proceedings of the 12th International Coral Reef Symposium, 9-13 July 2012, Cairns, QLD, Australia. Pp.1-5.

Anderson, K.D., Heron, S.F., and Pratchett, M.S. (2015). Species-specific declines in the linear extension of branching corals at a subtropical reef, Lord Howe Island. Coral Reefs, 34(2), 479-490.

Bauman, A.G., Baird, A.H., Burt, J.A., Pratchett, M.S., and Feary, D.A. (2014). Patterns of coral settlement in an extreme environment: the southern Persian Gulf (Dubai, United Arab Emirates). Marine Ecology Progress Series, 499, 115-126.

Beger, M., Sommer, B., Harrison, P.L., Smith, S.D., and Pandolfi, J.M. (2014). Conserving potential coral reef refuges at high latitudes. Diversity and Distributions, 20(3), 245- 257.

Bellwood, D.R., Hughes, T.P., Folke, C., and Nyström, M. (2004). Confronting the coral reef crisis. Nature, 429, 827-833

Bellwood, D.R., Hughes, T.P. and Hoey, A.S. (2006). Sleeping functional group drives coral-reef recovery. Current Biology, 16, 2434-2439.

Bonaldo, R.M., and Bellwood, D.R. (2008). Size-dependent variation in the functional role of the parrotfish Scarus rivulatus on the Great Barrier Reef, Australia. Marine Ecology Progress Series, 360, 237-244.

Bruno, J.F., and Selig, E.R. (2007). Regional decline of coral cover in the Indo-Pacific: timing, extent, and subregional comparisons. PLoS One, 2(8), e711.

Casey, J.M., Baird, A.H., Brandl, S.J., Hoogenboom, M.O., Rizzari, J.R., Frisch, A.J., Mirbach, C.E. and Connolly, S.R. (2017). A test of trophic cascade theory: fish and benthic assemblages across a predator density gradient on coral reefs. Oecologia, 183, 161-175.

Choat, J.H., van Herwerden, L., Robbins, W.D., Hobbs, J.P. and Ayling, A.M. (2006). A report on the ecological surveys conducted at Middleton and Elizabeth Reefs, February 2006. Report to the Australian Government Department of Environment and Heritage.

Clements, K.D., German, D.P., Piché, J., Tribollet, A. and Choat, J.H. (2017). Integrating ecological roles and trophic diversification on coral reefs: multiple lines of evidence identify parrotfishes as microphages. Biological Journal of the Linnean Society, 120(4), 729-751.

Page 59

Dulvy, N.K., Freckleton, R.P. and Polunin, N.V. (2004). Coral reef cascades and the indirect effects of predator removal by exploitation. Ecology Letters, 7(5), 410-416.

Eakin, C.M. (1996). Where have all the carbonates gone? A model comparison of calcium carbonate budgets before and after the 1982–1983 El Nino at Uva Island in the eastern Pacific. Coral Reefs, 15(2), 109-119.

Edgar, G.J., Stuart-Smith, R.D., Willis, T.J., Kininmonth, S., Baker, S.C., Banks, S., et al. (2014). Global conservation outcomes depend on marine protected areas with five key features. Nature, 506(7487), 216. doi:10.1038/nature13022

Edmunds, P.J., and Carpenter, R.C. (2001). Recovery of Diadema antillarum reduces macroalgal cover and increases abundance of juvenile corals on a Caribbean reef. Proceedings of the National Academy of Science, USA 98, 5067-5071.

Ferreira, D.E.L., Peret, A.C., and Coutinho, R. (1998). Seasonal grazing rates and food processing by tropical herbivorous fishes. Journal of Fish Biology, 53(sA), 222-235.

Gilmour, J.P., Smith, L.D., Heyward, A.J., Baird, A.H., and Pratchett, M.S. (2013). Recovery of an isolated coral reef system following severe disturbance. Science, 340(6128), 69-71.

Glynn, P.W., Wellington, G.M., and Birkeland, C. (1979). Coral reef growth in the Galapagos: limitation by sea urchins. Science, 203(4375), 47-49.

Graham, N.A., Wilson, S.K., Jennings, S., Polunin, N.V., Bijoux, J.P. and Robinson, J. (2006). Dynamic fragility of oceanic coral reef ecosystems. Proceedings of the National Academy of Sciences, 103(22), 8425-8429.

Graham, N.A., Spalding, M.D., and Sheppard, C.R. (2010). Reef shark declines in remote atolls highlight the need for multi-faceted conservation action. Aquatic Conservation: Marine and Freshwater Ecosystems, 20(5), 543-548.

Graham, N.A., Ainsworth, T.D., Baird, A.H., Ban, N.C., Bay, L.K., Cinner, J.E., et al. (2011). From microbes to people: tractable benefits of no-take areas for coral reefs. Oceanography and Marine Biology-an Annual Review, 49, 105-136.

Graham, N.A., Jennings, S., MacNeil, M.A., Mouillot, D., and Wilson, S.K. (2015). Predicting climate-driven regime shifts versus rebound potential in coral reefs. Nature, 518(7537), 94-97.

Harriott, V.J. (1992). Recruitment patterns of scleractinian corals in an isolated sub-tropical reef system. Coral Reefs, 11(4), 215-219.

Harriott, V.J. (1995). Is the crown- of- thorns starfish a threat to the reefs of Lord Howe Island? Aquatic Conservation: Marine and Freshwater Ecosystems, 5(3), 179-190.

Harriott, V.J. (1998). Preliminary report on the status of corals and crown-of-thorns starfish at Middleton Reef, 1998. Report to Environment Australia.

Harriot, V.J. (1999). Coral growth in subtropical eastern Australia. Coral Reefs, 18, 281- 291.

Page 60

Harrison, P.L., Dalton, S.J., and Carroll, A.G. (2011). Extensive coral bleaching on the world’s southernmost coral reef at Lord Howe Island, Australia. Coral Reefs, 30(3), 775-775.

Hobbs, J.-P.A., and Feary D. (2007). Monitoring the ecological status of Elizabeth and Middleton Reefs, February 2007. Report to the Australian Government Department of Environment and Water Resources.

Hobbs, J.-P.A., Choat, J.H., Robbins, W.D., van Herwerden, L. and Feary, D.A. (2008). Unique fish assemblages at world’s southernmost oceanic coral reefs, Elizabeth and Middleton Reefs, Tasman Sea, Australia. Coral Reefs 27, 15.

Hoegh-Guldberg, O., Mumby, P.J., Hooten, A.J., Steneck, R.S., Greenfield, P., Gomez, E., Harvell, C.D., Sale, P.F., Edwards, A.J., Caldeira, K., et al. (2007). Coral reefs under rapid climate change and ocean acidification. Science, 318, 1737-1742.

Hoey, A.S., and Bellwood, D.R. (2008). Cross-shelf variation in the role of parrotfishes on the Great Barrier Reef. Coral Reefs, 27, 37-47.

Hoey, A.S., and Bellwood, D.R. (2009). Limited functional redundancy in a high diversity system: single species dominates key ecological process on coral reefs. Ecosystems, 12, 1316-1328.

Hoey, A.S., and Bellwood, D.R. (2010a). Cross-shelf variation in browsing intensity on the Great Barrier Reef. Coral Reefs, 29: 499-508.

Hoey, A.S., and Bellwood, D.R. (2010b). Among-habitat variation in herbivory on Sargassum spp. on a mid-shelf reef in the northern Great Barrier Reef. Marine Biology, 157(1), 189-200.

Hoey, A.S., and Bellwood, D.R. (2011). Suppression of herbivory by macroalgal density: a critical feedback on coral reefs? Ecology Letters, 14: 267-273.

Hoey, A.S., and Pratchett, M.S. (2017). Review of research and monitoring relevant to natural values in Australia’s Commonwealth Marine Reserves. Report to the Department of the Environment and Energy. 87pp.

Hoey, A.S., Brandl, S.J., and Bellwood, D.R. (2013). Diet and cross-shelf distribution of rabbitfishes (f. Siganidae) on the northern Great Barrier Reef: implications for ecosystem function. Coral Reefs, 32(4), 973-984.

Hoey, A.S., Pratchett, M.S., Johansen, J., and Hoey, J. (2014). 2014 Marine ecological survey of Elizabeth and Middleton reefs, Lord Howe Commonwealth Marine Reserve. Report to the Department of the Environment. 99pp.

Hoey, A.S., Pratchett, M.S., and Cvitanovic, C. (2011). Low rates of herbivory and coral recruitment undermines the potential resilience of coral reef assemblages at Lord Howe Island. PLoS ONE, 6(10), e25824

Hoey, A.S., Berumen, M.L., Bonaldo, R.M., Burt, J.A., Feary, D.A. et al. (2018). The Ecology of Parrotfishes in Marginal Reef Systems. In: Hoey, A.S. & Bonaldo, R. M. (Eds). Biology of Parrotfishes (pp. 276-301). CRC Press.

Hughes, T.P. (1994). Catastrophes, phase shifts, and large-scale degradation of a Caribbean coral reef. Science, 265(5178), 1547-1551.

Page 61

Hughes, T.P., Baird, A.H., Bellwood, D.R., Card, M., Connolly, S.R., Folke, C., et al. (2003). Climate change, human impacts, and the resilience of coral reefs. Science, 301(5635), 929-933.

Hughes, T.P., Rodrigues, M.J., Bellwood, D.R., Ceccarelli, D., Hoegh-Guldberg, O., McCook, L. et al. (2007). Phase shifts, herbivory, and the resilience of coral reefs to climate change. Current Biology, 17(4), 360-365.

Hughes, T. P., Kerry, J. T., Álvarez-Noriega, M., Álvarez-Romero, J. G., Anderson, K. D., Baird, A. H. et al. (2017a). Global warming and recurrent mass bleaching of corals. Nature, 543(7645), 373.

Hughes, T.P., Barnes, M.L., Bellwood, D.R., Cinner, J.E., Cumming, G.S., Jackson, J.B., et al. (2017b). Coral reefs in the Anthropocene. Nature, 546(7656), 82.

Hughes, T.P., Anderson, K.D., Connolly, S.R., Heron, S.F., Kerry, J.T., Lough, J.M., et al. (2018). Spatial and temporal patterns of mass bleaching of corals in the Anthropocene. Science, 359(6371), 80-83.

Jackson, J.B., Kirby, M.X., Berger, W.H., Bjorndal, K.A., Botsford, L.W., Bourque, B.J., et al. (2001). Historical overfishing and the recent collapse of coastal ecosystems. Science, 293(5530), 629-637.

Last, P.R. and Stevens, J.D. (1994). Sharks and Rays of Australia. Commonwealth Scientific and Industrial Research Organization, Australia. 513 pp.

Lewis, S.M. (1986). The role of herbivorous fishes in the organization of a Caribbean reef community. Ecological Monographs, 56(3), 183-200.

Lokrantz, J., Nyström, M., Thyresson, M., and Johansson, C. (2008). The non-linear relationship between body size and function in parrotfishes. Coral Reefs, 27(4), 967- 974.

Lourey, M.J., Ryan, D.A.J., and Miller, I.R. (2000). Rates of decline and recovery of coral cover on reefs impacted by, recovering from and unaffected by crown-of-thorns starfish Acanthaster planci: a regional perspective of the Great Barrier Reef. Marine Ecology Progress Series, 196, 179–186.

Liu, G., Strong, A.E., & Skirving, W. (2003). Remote sensing of sea surface temperatures during 2002 Barrier Reef coral bleaching. Eos, Transactions American Geophysical Union, 84(15), 137-141.

McClanahan, T.R., and Shafir, S. H. (1990). Causes and consequences of sea urchin abundance and diversity in Kenyan coral reef lagoons. Oecologia, 83, 362-370.

McClanahan, T.R., Baird, A.H., Marshall, P.A., and Toscano, M.A. (2004). Comparing bleaching and mortality responses of hard corals between southern Kenya and the Great Barrier Reef, Australia. Marine Pollution Bulletin, 48(3-4), 327-335.

Mumby, P.J. (2006). The impact of exploiting grazers (Scaridae) on the dynamics of Caribbean coral reefs. Ecological Applications, 16(2), 747-769.

Mumby, P.J., and Steneck, R.S. (2008). Coral reef management and conservation in light of rapidly evolving ecological paradigms. Trends in Ecology and Evolution, 23, 555- 563.

Page 62

Nishiguchi, S., Wada, N., Yamashiro, H., Ishibashi, H., and Takeuchi, I. (2018). Continuous recordings of the coral bleaching process on Sesoko Island, Okinawa, Japan, over about 50 days using an underwater camera equipped with a lens wiper. Marine Pollution Bulletin, 131, 422-427.

Oliver, T.A., and Palumbi, S.R. (2011). Do fluctuating temperature environments elevate coral thermal tolerance?. Coral Reefs, 30(2), 429-440.

Oxley, W.G., Ayling, A.M., Cheal, A.J. and Osborne, K. (2004). Marine surveys undertaken in the Elizabeth Middleton Reefs Marine National Nature Reserve, December 2003. Report to the Australian Government Department of Environment and Heritage.

Pandolfi, J.M., Bradbury, R.H., Sala, E., Hughes, T.P., Bjorndal, K.A., Cooke, R.G., McArdle, D., McClenachan, L., Newman, M.J.H., Paredes, G., Warner, R.R. and Jackson, J.B.C. (2003). Global Trajectories of the Long-Term Decline of Coral Reef Ecosystems. Science, 301, 955-958.

Polunin, N.V.C., & Klumpp, D.W. (1992). Algal food supply and grazer demand in a very productive coral-reef zone. Journal of Experimental Marine Biology and Ecology, 164(1), 1-15.

Pratchett, M.S. (2005). Dynamics of an outbreak population of Acanthaster planci at Lizard Island, northern Great Barrier Reef (1995–1999). Coral Reefs, 24(3), 453-462.

Pratchett, M.S. (2010). Changes in coral assemblages during an outbreak of Acanthaster planci at Lizard Island, northern Great Barrier Reef (1995–1999). Coral Reefs, 29(3), 717-725.

Pratchett, M.S., Hobbs, J-P.A., Hoey, A.S., Baird, A.H., Ayling, A.M., Gudge, S., and Choat, J.H. (2011). Elizabeth and Middleton Reefs Marine National Nature Reserve: Marine Survey 2011. Report for Department of Sustainability, Environment, Water, Population and Communities. 84 pages.

Pratchett, M.S., Caballes, C., Rivera-Posada, J.A., and Sweatman, H.P.A. (2014a). Limits to understanding and managing outbreaks of crown-of-thorns starfish (Acanthaster spp.). Oceanography and Marine Biology: An Annual Review, 52: 133-200.

Pratchett, M.S., Hoey, A.S., Cvitanovic, C., Hobbs, J.P.A., and Fulton, C.J. (2014b). Abundance, diversity, and feeding behavior of coral reef butterflyfishes at Lord Howe Island. Ecology and Evolution, 4(18), 3612-3625.

Pratchett, M.S., Anderson, K.D., Hoogenboom, M.O., Widman, E., Baird, A.H., Pandolfi, J. M., et al. (2015). Spatial, temporal and taxonomic variation in coral growth— implications for the structure and function of coral reef ecosystems. Oceanography and Marine Biology: An Annual Review, 53, 215-295.

Pratchett, M.S., Caballes, C.F., Wilmes, J.C., Matthews, S., Mellin, C., Sweatman, H., et al. (2017). Thirty years of research on crown-of-thorns starfish (1986–2016): Scientific advances and emerging opportunities. Diversity, 9(4), 41.

Purcell, S.W. (2014). Value, Market Preferences and Trade of Beche-De-Mer from Pacific Island Sea Cucumbers. PloS ONE, 9(4), e95075.

Page 63

Purcell, S.W., Polidoro, B.A., Hamel, J.-F., Gamboa, R.U., and Mercier, A. (2014). The cost of being valuable: predictors of extinction risk in marine invertebrates exploited as luxury seafood. Proceedings of the Royal Society, London B, 281

Rasher, D.B., Hoey, A.S., and Hay, M.E. (2013). Consumer diversity interacts with prey defenses to drive ecosystem function. Ecology, 94(6), 1347-1358.

Rasher, D.B., Hoey, A.S., and Hay, M.E. (2017). Cascading predator effects in a Fijian coral reef ecosystem. Scientific Reports, 7(1), 15684.

Richardson, L.E., Graham, N.A., Pratchett, M.S., Eurich, J.G. and Hoey, A.S. (2018). Mass coral bleaching causes biotic homogenization of reef fish assemblages. Global Change Biology. Doi:10.1111/gcb.14119

Riegl, B. (2003). Climate change and coral reefs: different effects in two high-latitude areas (Arabian Gulf, South Africa). Coral Reefs, 22(4), 433-446.

Rizzari, J.R., Frisch, A.J., Hoey, A.S., and McCormick, M.I. (2014). Not worth the risk: apex predators suppress herbivory on coral reefs. Oikos, 123(7), 829-836.

Robbins, W.D., Hisano, M., Connolly, S.R., and Choat, J.H. (2006). Ongoing collapse of coral-reef shark populations. Current Biology, 16(23), 2314-2319.

Russ, G.R. (1984). Distribution and abundance of herbivorous grazing fishes in the central Great Barrier Reef. I. Levels of variability across the entire continental shelf. Marine Ecology Progress Series, 20(1), 23-34.

Russ, G.R. (2003). Grazer biomass correlates more strongly with production than with biomass of algal turfs on a coral reef. Coral Reefs, 22(1), 63-67.

Russ, G.R., and Alcala, A.C. (2010). Decadal-scale rebuilding of predator biomass in Philippine marine reserves. Oecologia, 163(4), 1103-1106.

Rylaarsdam, K.W. (1983). Life histories and abundance patterns of colonial corals on Jamaican reefs. Marine Ecology Progress Series, 13, 249–260.

Smith, L.D., Gilmour, J.P., and Heyward, A.J. (2008). Resilience of coral communities on an isolated system of reefs following catastrophic mass-bleaching. Coral Reefs, 27, 197-205

Stuart-Smith, R.D., Ceccarelli, D., and Edgar, G.J. 2017 Biodiversity Survey of the Temperate East Coast Commonwealth Marine Reserve Network: Elizabeth & Middleton Reefs, Lord Howe Island & Norfolk Island. Report to Parks Australia, Department of the Environment. 59pp.

Sussman, M., Willis, B. L., Victor, S., and Bourne, D. G. (2008). Coral pathogens identified for white syndrome (WS) epizootics in the Indo-Pacific. PLoS One, 3(6), e2393.

Sweatman, H. (2008). No-take reserves protect coral reefs from predatory starfish. Current Biology, 18, 598-599.

Sweet, M., and Bythell, J. (2012). Ciliate and bacterial communities associated with White Syndrome and Brown Band Disease in reef- building corals. Environmental Microbiology, 14(8), 2184-2199.

Page 64

Trapon, M.L., Pratchett, M.S., and Hoey, A.S. (2013a). Spatial variation in abundance, size and orientation of juvenile corals related to the biomass of parrotfishes on the Great Barrier Reef, Australia. PLoS ONE, 8(2), e57788.

Trapon, M.L., Pratchett, M.S., Hoey, A.S., and Baird, A.H. (2013b). Influence of fish grazing and sedimentation on the early post-settlement survival of the tabular coral Acropora cytherea. Coral Reefs, 32(4), 1051-1059.

Uthicke, S., Schaffelke, B. and Byrne, M. (2009). A boom–bust phylum? Ecological and evolutionary consequences of density variations in echinoderms. Ecological Monographs, 79(1), 3-24.

Valentine, J.P., and Edgar, G.J. (2010). Impacts of a population outbreak of the urchin Tripneustes gratilla amongst Lord Howe Island coral communities. Coral Reefs 29, 399-410.

Wilkinson, C. (2008) Status of coral reefs of the world: 2008. Global Coral Reef Monitoring Network and Reef and Rainforest Research Centre, Townsville, Australia.

Williams, I.D., Polunin, N.V. and Hendrick, V.J. (2001). Limits to grazing by herbivorous fishes and the impact of low coral cover on macroalgal abundance on a coral reef in Belize. Marine Ecology Progress Series, 222, 187-196.

Wilson, S.K., Graham, N.A.J. and Polunin, N.V.C. (2007). Appraisal of visual assessments of habitat complexity and benthic composition on coral reefs. Marine Biology 151, 1069-1076.

Wismer, S., Hoey, A.S., and Bellwood, D.R. (2009). Cross-shelf benthic community structure on the Great Barrier Reef: relationships between macroalgal cover and herbivore biomass. Marine Ecology Progress Series 376, 45-54.

Page 65