Water Mass connectivity and mixing along the southern margin of Australia: hydrographic and stable isotope analyses

Laura Ellen Richardson

A thesis submitted for the degree of Doctor of Philosophy Of The Australian National University

April 2015 Statement of Authorship

Unless otherwise acknowledged, this thesis is my own original work. It would not have been possible without the intellectual and scientific guidance of John Middleton, Bradley Opdyke, Noel James and Kurt Kyser, who supervised the research and are co-authors on the three manuscripts (chapters 2, 3 and 4).

Laura Richardson April 2015 Acknowledgements

Firstly, thanks to my supervisor John Middleton at SARDI Aquatic Sciences. John you allowed me to follow the path I wanted to take with this project. I am eternally grateful for the opportunity to be part of the physical oceanography group and learn so much, and to be able to go out to sea and collect my own data. The cruises were such a great experience! I thank you for providing the funding to send me to Canada to run the isotope samples, without which I would not have been able to go. You have been very supportive through the whole process both personally and scientifically, even when I needed to move to Kalgoorlie and then Canada. I think very fondly of my time at SARDI, it was a great environment to work in and I learnt a great deal. I hope I was able to return the favour and teach you a little about isotopes too!

Thanks to my Canadian advisors Noel James and Kurt Kyser at Queen’s University. Noel you have always been so supportive, believing in me and directing me back on the water mass path when the physical oceanography around me was getting too much! Thanks for listening as I went through my ideas and data with you, and for editing my manuscripts so thoroughly. Kurt I am so grateful for having access to the QF1R stable isotope lab - thank you for giving me the chance to run so many samples. Your insight into isotope processes has been so valuable and I have learnt so much from you.

A big thank you to Kerry Klassen and April Vuletich in the stable isotope lab for helping me run the machines, especially all the troubleshooting that needed to be done. I appreciate your time and help. Thanks very much to Charles James for all your patience teaching me the basics of Matlab and helping me with scripts, and for helping me define oceanography terms while I was correcting my thesis. I really appreciate you helping me remotely to make figures when my scripts were not working! And thanks to Mark Doubell for helping me with scripts when Charles’ explanations were a bit too technical for me!

Last but not least, thank you to my supervisor on the ground, Brad Opdyke. Things didn’t really go as planned but thanks for being so supportive and letting me change what I wanted to study. I really appreciate all your help remotely, and for being a friendly face when I came to visit. You’ve been a wonderful mentor since undergrad. You were central to me finding what I loved to study and have been an important part of the journey since then.

1 would like to thank my wonderful husband Jonathan, for your continual support and guidance; for listening to me formulate ideas, problem solve and share my enthusiasm for the project, and for your help with all things technical! Thank you so much for being there when I needed support, and for working so hard to give me the chance to do what I needed to do. I would not have gotten to this point without you. Many thanks to my Mum and Dad, for always believing in me and supporting the path I have taken without question or judgment. I am so lucky to have such a strong foundation and constant encouragement. Thank you to my Aunty Riet and Uncle Bruce for being so interested and invested in my schooling! It has been a long road but I have finally gotten here! Thanks Aunty Riet for all your writing and editing help. And thank you to the rest of my family and friends who have supported me through this process.

Finally to my son Liam, you made this process so much more challenging and interesting! I’m so glad you were part of this journey. Abstract

This study is the first to characterise the hydrographic properties and depth range of the Flinders Current and confirm its influence on shelf ecosystems of the Kangaroo Island upwelling region. Four water masses are identified in the top 1000 meters water depth (mwd) from Cape Leeuwin to Tasmania, using hydrochemistry and stable isotopes of seawater. Three water masses are identified from previous literature on the southeast Indian Ocean: Subtropical Surface Water (STSW), Tasmanian Subantarctic Mode Water (TSAMW) and Tasmanian Intermediate Water (TIW), and one is newly identified and named: South Australian Basin Central Water (SABCW). STSW is transported east by the Leeuwin Current system and is modified by heating and evaporation along the subtropical continental shelf. SABCW is formed at the subtropical front within the South Australian Basin at -40°S, TSAMW is formed within the Subantarctic Zone southwest of Tasmania, and TIW is formed from mixing of two different types of Antarctic Intermediate Water west of Tasmania.

The Flinders Current transports SABCW, TSAMW and TIW west along the Australian continental slope. The top surface of SABCW delineates the interface between subantarctic water transported by the Flinders Current and subtropical water transported by the Leeuwin Current system. This interface is typically -300 mwd during winter and -250 mwd during summer, but can be as shallow as 150 mwd during summer in the Kangaroo Island upwelling region and off western Tasmania. Stable isotope values show these water masses continue north along the Western Australian slope, identifying connectivity between the Flinders Current and Leeuwin Undercurrent.

Deep upwelling events in the Kangaroo Island upwelling region source SABCW from depths of 300 m or more, which is the first evidence that upwelling supplies Flinders Current water to shelf ecosystems. Stable isotopes of seawater identify the formation of a mixed water mass as SABCW mixes with STSW on the shelf. Spatial distribution of this water mass suggests that upwelled water is transported west towards Eyre Peninsula and north into the mouth of Spencer Gulf, and vertical mixing allows upwelled nutrients to be brought into the photic zone to be utilised by primary producers. Strong upwelling events during February and March 2008 and February and March 2010 recorded temperatures/salinities as low as 10.4°C/34.85, and NOx/phosphate concentrations as high as 13.35/0.94 pmol/L, on the shelf. New results for nutrients show average values of NOx and phosphate during months of strong upwelling to be 6.1 times and 4.6 times greater, respectively, than during winter months, and that upwelled water can have nutrient concentrations up to 90 times higher than those in summer surface waters, which is higher than values recorded previously for the Bonney Coast. Upwelled water was also low in silicate, a signature of Southern Ocean water masses, which has implications for phytoplankton community structure and diatom abundance on the shelf. Identifying nutrient signatures of upwelled water, as well as water mass interactions during upwelling events, has implications for mixing of nutrient-rich upwelled waters with oligotrophic surface waters, a situation that supports greater levels of primary productivity on the shelf. Table of Contents

STATEMENT OF AUTHORSHIP...... II

ACKNOWLEDGEMENTS...... Ill

ABSTRACT...... V

LIST OF FIGURES...... XI

LIST OF TABLES...... XIII

LIST OF ABBREVIATIONS...... XIV

CHAPTER 1. INTRODUCTION...... 1

1.1 M o t iv a t io n a n d sc o pe of t h e s is ...... 3

1.2 T hesis o u t l in e a n d key q u e s t io n s ...... 4

1.3 M e t h o d o l o g y Ov e r v ie w ...... 7

Hydrochemistry data...... 7

Stable isotope data...... 7

Stable isotope analyses...... 10

1.4 Ref er e n c e s ...... 11

CHAPTER 2. SHALLOW WATER MASSES AND THEIR CONNECTIVITY ALONG THE SOUTHERN

MARGIN OF AUSTRALIA...... 15

ABSTRACT...... 15

2.1. INTRODUCTION...... 16

2.1.1 Stabl e Isot op es of Se a w a t e r ...... 20

2.2. M ETHO DS...... 21

2.3. RESULTS AND DISCUSSION...... 23

2.3.1 Da ta An a l y s is ...... 23

2.3.2 Su b t r o pic a l Surfa ce W ater (STSW)...... 26

Origins and spatial distribution...... 26

Nutrients...... 31

2.3.3 So u t h Au s tr al ia n Bas in Cen tr al W ater (SABCW)...... 32

Origins and spatial distribution...... 32

Nomenclature...... 35

Nutrients...... 36

vii 2.3.4 Ta s ma n ia n Sub an ta r ct ic M o d e W at er (TSAMW)...... 37

Origins and spatial distribution...... 37 Nutrients...... 38

2.3.5 Ta s ma n ia n In t er med ia te W at er (T IW )...... 38

Origins and spatial distribution...... 38

Nutrients...... 39

2.3.6 Sta bl e Is o t o pe Ana ly s is ...... 40

Subtropical Surface Water...... 41 South Australian Basin Central Water...... 43

Tasmanian Subantarctic Mode Water...... 44

Tasmanian Intermediate Water...... 45

2.3.7 Reg io n al Connectivity ...... 46

Leeuwin Undercurrent connectivity...... 49

Tasman Sea influence...... 50

2.4. CONCLUSIONS...... 51

2.5. REFERENCES...... 55

CHAPTER 3. WATER MASSES AND THEIR SEASONAL VARIATION ON THE LINCOLN SHELF,

SOUTH AUSTRALIA...... 64

ABSTRACT...... 64

3.1. INTRODUCTION...... 65

3.1.1. Reg ion al cir c ul ati on a n d w at e r ma s s e s ...... 66

3.1.2. Is o t o pes in subtr op ical r e g io n s ...... 68

3.2. M ETHO DS...... 69

3.3. SUMMER WATER MASSES...... 72

3.3.1. Introduction ...... 72

3.3.2. Sl o pe W a t e r ...... 76

Definition and origins...... 76

Spatial distribution...... 76

3.3.3. Subt ropical Su r f ac e W a t e r ...... 77

Definition and origins...... 77

Spatial distribution...... 77

3.3.4. Evapo ra ted W a t e r ...... 78 viii Definition and origins...... 78 Spatial distribution...... 78

3.B.5. M ix ed Sl o pe W a t e r ...... 78

Definition and origins...... 78

Spatial distribution...... 78

3.3.6. Coo led Evapora ted W at er ...... 79

Definition and origins...... 79 Spatial distribution...... 79

3.3.7. Imp li c a ti o n s f o r m ix in g ...... 80

3.4. WINTER WATER MASSES...... 81

3.4.1. Introduction ...... 81

3.4.2. Sl o pe W a t e r ...... 84

Definition...... 84 Spatial distribution...... 85

3.4.3. Sub tropical Su r f ac e W a t e r ...... 85

Definition...... 85 Spatial distribution...... 85

3.4.4. Eva po ra ted W at er ...... 85

Definition...... 85 Spatial distribution...... 86 Spencer Gulf Outflow...... 86

3.4.5. Impl ic a ti o ns f o r m ix in g ...... 87

3.5. DISCUSSION...... 87

3.5.1. Tem po ra l a n d s eas on al v a r ia t io n ...... 87

3.5.2. Co mp a r is o n w it h pre vi ou s wate r mass s t u d ie s ...... 89

3.5.3. Co mp a r is o n w it h pre vi ou s is o t o pe s t u d ie s ...... 90

3.6. CONCLUSIONS...... 91

3.7. REFERENCES...... 94

CHAPTER 4. UPWELLING CHARACTERISTICS AND NUTRIENT ENRICHMENT OF THE

KANGAROO ISLAND UPWELLING REGION, SOUTH AUSTRALIA...... 100

ABSTRACT...... 100

4.1. INTRODUCTION...... 101

ix 4.2. METHODS 104

4.3. WATER MASS CONTEXT...... 105

4.4. RESULTS...... 107

4.4.1. Upwelling Ch a r a ct e ri st ic s ...... 107

2008 Upwelling Season...... I l l

2010 Upwelling Season...... 114

4.4.2 Nut r ie n ts a n d Up w e l l in g ...... 117

Nutrient relationships...... 118

Seasonal states...... 120

4.5. DISCUSSION...... 123

4.5.1 Dept h o f u p w e l l in g ...... 123

4.5.2 Upwelling me c h a n is ms a n d v a r ia t io n bet we en ye ar s ...... 125

4.5.3 Nutrien t e n r ic h me n t du ri ng up wel ling e v en ts ...... 127

4.5.4 Sil ic at e signature a n d s o u r c e o f upwe ll ed w a t e r ...... 129

4.6. CONCLUSIONS...... 131

4.7. REFERENCES...... 133

CHAPTER 5. RESEARCH CONCLUSIONS...... 138

5.1 KEY FINDINGS...... 139

5.1.1 Regional water masses and Flinders Current connectivity...... 139

5.1.2 Lincoln Shelf water masses...... 141

5.1.3 Upwelling characteristics o f the Kangaroo Island upwelling region...... 144

5.2 Resear c h Imp l ic a t io n s ...... 146

5.2.1 Identification o f the Flinders Current...... 146

5.2.2 Use of stable isotopes in shelf water mass studies...... 146

5.2.3 Identifying upwelling characteristics...... 147

5.3 Fut u re re se ar c h ...... 148

5.4 Co n c l u d in g s t a t e me n t ...... 149

5.5 Refe r en c es ...... 150

APPENDIX A - DATA FILE FOR THE SOUTHERN MARGIN ...... 151

APPENDIX B - IN-SITU TEMPERATURE AND POTENTIAL DENSITY...... 158

APPENDIX C - SAIMOS DATA FILE FOR THE LINCOLN SHELF...... 161

x List of Figures

Fig u r e 1.1. M aj o r ocean c u rr ent s in fl u en c in g t he southe rn ma r g in o f Au s t r a l ia ...... 2

Fig u r e 1.2. Upw e ll in g region in So u th A ust ra lia betwee n t he eas t er n Gr eat A us tra lian Bigh t a n d t h e

Bon n e v Co a s t ...... 3

Fig u r e 1.3. h o w hydro lo gic pr o c esses af fe c t st ab l e iso to pe composit ions ...... 8

Fig u r e 2.1. Re g io n a l cir c u la tio n o f south ern A u s t r a l ia ...... 17

Fig u r e 2.2. sta ti on loca tio ns f o r n u t r ie n t a n d st abl e iso t o pe sa mples fr om RVSo u t h e r n Sur ve yo r

VOYAGES DURING 2 0 1 0 ...... 22

Fig u r e 2.3. Temp er a tu r e -sal inity plo t f o r sel ec t regions a lo n g t h e southern m a r g in ...... 25

Fig u r e 2.4. Avera ged w in t e r pr o per t ie s a lo n g the southe rn ma r g in , 110°E t o 147°E...... 28

Fig ur e 2.5. Avera ged s u mme r pr o per t ie s a lo n g th e s out he rn ma r g in , 110°E to 147°E...... 29

Fig u r e 2.6. Aver aged y ear l y n u tr ie n t c on c en t r at io n s a l o n g the southern ma r g in , 115°Eto 147°E...... 30

Fig u r e 2.7. Nu tri en t pr o f il es by w at e r mas s f o r s ta tion s s am pl ed d u r in g th e A ug u s t 2010 Sout he rn

Surv eyo r vo yage a lo n g th south ern ma r g in ...... 31

Fig u r e 2.8. W int er sc he ma ti c a n d s u mme r l a t it u d in a l sal inity se ct io n o f w a t e r mas se s sou th o f

A u s t r a l ia ...... 34

Fig u r e 2 .9 .ö180-ö2H relation sh ip f o r all re gi on a l iso t o pe s a mp l e s ...... 42

Fig u r e 2.10. Isotope -sal inity rel ationships f o r al l r e gi on a l iso t o pe s a mp l e s ...... 43

Fig ur e 2.11. Isotope -de pth pr o f il es f o r al l r e gi on a l iso t o pe s a mp l e s ...... 45

Fig u r e 2.12. Sc h e ma t ic o f cu rr en ts a n d wa t e r mass es south o f A u s t r a l ia ...... 48

Fig ure 3.1. Lin c o ln Sh el f st ud y area w it h lo c a l c u r r e n t s ...... 65

Fig ur e 3.2. Lin c ol n Shel f st ud y area w it h sam pl e l o c a t io n s ...... 71

Fig u r e 3.3. Su mme r t e mp e r at u r e -sal inity pl o t f o r the Lin col n Sh e l f ...... 73

Fig u r e 3.4. Su mme r ö 2H v s ö 180 a n d tem p e r atu r e v s sal inity o f iso t o pe samples by w at e r ma s s ...... 74

Fig u r e 3.5. Sc h e ma t ic o f s u mme r s h el f wa t e r mas s es d u r in g pe ri od s o f u p w e l l in g ...... 75

Fig u r e 3.6. Bo t t o m pl o t s o f te mp er atu re a n d sal in ity f o r Feb ru ar y 2010 s h o w in g u p w e l l in g ...... 77

Fig u r e 3.7. W int er te mp er atu re vs sal inity plo t f o r Lin c ol n Sh e l f ...... 82

Fig u r e 3.8: W int er ö 2H vs ö 180 a n d tem p e r atu r e v s sal inity o f iso t o pe samples by w at e r m a s s ...... 83

Fig u r e 3.9. Su mme r a n d w in t e r valu es f o r ö 2H vs ö 180 a n d ö 2H vs sal inity o n t he Lin c ol n Sh e lf ...... 84

Fig u r e 3.10. Bo t t o m pl o t s o f sal inity a n d de ns ity f o r Oc t o be r 2008 s h o w in g Spen c er Gu l f o u t f l o w .....86

Fig u r e 4.1. St a t io n lo ca ti on s f o r n u t r ie n t sampl es o n t he Lin col n Sh e l f ...... 104

Fig u r e 4.2. Tempera tur e vs sal inity o f SAIMOS d a ta f o r t he Lin col n Sh e l f ...... 108

Fig u r e 4.3. Tempera tur e vs sal inity by seasonal s t at e o f shel f wa t e r fr o m d ep th s o f 100-120 m ...... 109

Fig u r e 4.4. Pr o f il es o f tem p e r atu r e a n d sal inity c o mp a r in g u p w e ll in g a n d n o n -u p w e ll in g mo n t h s ...... I l l

Fig u r e 4.5. Sea Su r f ac e Tempera tur e ima g e o f a strong u p w e ll in g event on t h e 14ih Feb rua ry 200 8..... 112

Fig u r e 4.6. Bo t t o m pl o t s o f te mper atu re a n d sal in ity f o r mo n t h s o f st ro ng u p w e ll in g d u r in g 2008... 113

Fig u r e 4.7. Bo t t o m sal inity pl o t s f o r Nov e mb e r 2008 a n d No v e mb er 2 0 0 9 ...... 115

xi Fig ur e 4.8. Sea Su r f ac e Temp er a tu r e ima g e o f a st ro ng u p w e ll in g event on t he 3rd M ar c h 2010...... 116

Fig ur e 4.9. Bo t t o m pl o t s o f t e mp e r at u r e a n d sal inity f o r mo n t h s o f strong u p w e ll in g d u r in g 2010.... 117

Fig ur e 4.10. Nut r ie n t r e la tion sh ips by sea so na l state ...... 119

Fig ur e 4.11. Si* pr o f il es by s ea so n al s t a t e ...... 121

Fig ur e 4.12. Dept h pr o f il es o f NOx, phosphate a n d s il ic at e by seaso na l state ...... 122

Fig ur e 4.13. M ean w in d st r ess f o r Neptu ne Is la n d , Sou th A us tralia d u r in g s u mme r a n d w in t e r ...... 126

Figur e 4.14. W eekl y ave r a ge d w in d st r ess f o r Nep tu ne Is l a n d ...... 127

Figur e 5.1. Sc h e ma t ic o f curr ents a n d w at e r mas s es south o f A u s t r a l ia ...... 140

Figur e 5.2. Averaged s u mme r te mp e r at u r e a n d salinity se ct io ns a l o n g th e southe rn ma r g in fr om 110°E

TO 147°E...... 140

Fig ur e 5.3. Sc h e ma t ic o f s u mme r w at e r mas se s on th e Linc oln Sh e l f ...... 143

Fig ur e 5.4. Su mma r y o f w a t e r pr o per t ie s on th e Lin coln Shel f by seas ona l s ta te : te mper atu re vs sal inity

AND NOx VS PHOSPHATE 145 List of Tables

Tab le 2.1. Pr o per t ies o f w at e r mas s es a lo n g t h e so uth er n ma r g in ...... 24

Tab le 3.1: Su mme r w at e r mass pr o per t ie s on t h e Linc ol n Sh e l f ...... 75

Tab le 3.2: W int er wa t e r mass pr o per t ies on t h e Lin c ol n Shelf ...... 84

Tab le 4.1: Dup lic at e o f Tab le 3 .1 ...... 107

Tab le 4.2. Temp er a tu r e -sa linity pr o per t ie s o f stro ng u p w e ll in g ev en ts o n t h e Linc oln Sh e l f ...... 110

Tab le 4.3. Ave ra g e values o f nutr ien ts by seasonal s t at e f o r th e Lin c o ln Sh e l f ...... 118

xiii List of Abbreviations

Currents

LC - Leeuwin Current

SAC - South Australian Current

ZC - Zeehan Current

FC - Flinders Current

EAC - East Australian Current

Water Masses

STSW - Subtropical Surface Water

SABCW - South Australian Basin Central Water

SAMW - Subantarctic Mode Water

TSAMW - Tasmanian Subantarctic Mode Water

AAIW - Antarctic Intermediate Water

TIW - Tasmanian Intermediate Water NIIW-Northwest Indian Intermediate Water

TSW - Tropical Surface Water SICW - South Indian Central Water

ICW - Indian Central Water

Lincoln Shelf Water Masses

SW - Slope Water

MSW - Mixed Slope Water

EW - Evaporated Water

CEW - Cooled Evaporated Water

Locations/Features

GAB - Great Australian Bight

N WC - North West Cape, Western Australia

STF - Subtropical Front

SAF - Subantarctic Front

SAZ - Subantarctic Zone xiv Nutrients

NO3 - Nitrate

N 0X - Nitrate + nitrite

PO4 - Phosphate

SiOi - Silicate

Si* - the relative abundance of silicate to nitrate (Si* = [SiÜ2] - [NO3]) Chapter 1. Introduction

The southern margin of Australia is an important region for both surface and intermediate water masses, yet it has been minimally studied compared to the eastern, western and northern margins of Australia. The southern margin is one of the world’s largest latitude-parallel shelf and slope regions (James et al., 1994), which hosts an extensive cool-water carbonate province of ecological and environmental importance, as well as an upwelling region that supports several major Australian Commonwealth and State fisheries (McLeay et al., 2003; Ward et al., 2006). Only a couple of localised water mass studies have been undertaken for the shelf, however, and no studies have focused on water masses deeper than 200 m for the continental slope. Instead, understanding of the regional oceanography has come from modelling of the shelf and slope currents (Figure 1.1; e.g. Middleton and Cirano, 2002; Middleton and Platov, 2003; Cirano and Middleton, 2004). Despite this, the southern margin of Australia is of vital importance to intermediate water mass transfer on a global scale; it is a possible link between the Pacific and Indian Oceans (Speich et al., 2002), and is a transfer pathway of Subantarctic Mode Water, which forms in the Subantarctic Zone and is transported north to ventilate the southern hemisphere subtropical gyres (McCartney, 1982).

Subantarctic Mode Water is transported by the Flinders Current, a northern boundary current that flows west along the southern margin with a core depth of 400-600 m and speeds of up to 20 cm/s (Middleton and Bye, 2007). This current is poorly understood, having been defined using numerical models with minimal and/or conflicting observational data. The current has not been defined hydrographically, nor have the water masses it transports been defined in regard to depth distribution and seasonal variation. In addition, determining the connectivity of the southern margin with the eastern and western margins of Australia via the Flinders Current is essential to understanding the link between the Pacific and Indian Oceans. This current may provide a secondary pathway of intermediate water in addition to the

1 Indonesian Throughflow north of Australia, which in comparison has been extensively studied.

AUSTRALIA Leeuwin Current

South Australian ^ -.C u rre n t Spencer Gulf r" outflow

Zeehan Current

Subantarctic Zone West wind drift —- Subantarctic Mode V Tasman Outflow Water formation / N

110°E 120°E 130°E 140°E 150°E Longitude (°E)

Figure 1.1. Major ocean currents influencing the southern margin o f Australia. Surface currents - solid line, deeper currents - dashed line. STF = Subtropical Front. Question marks denote uncertainty with regional connectivity.

Determining the hydrography and depth distribution o f the Flinders Current is also imperative to understanding the influence o f Southern Ocean water on shelf ecosystems in the upwelling region between the eastern Great Australian Bight and the Bonney Coast, South Australia (Figure 1.2). This known upwelling region is ecologically and economically important, but has also been minimally studied. Current knowledge o f upwelling is based on sea surface temperature (SST) imagery with only several hydrographic and biological studies, (e.g. Lewis, 1981; Schahinger, 1987; McClatchie et al., 1997; van Ruth et al., 2010a,b). Major questions still remain. Establishing the influence o f either Southern Ocean water from the Flinders Current or tropical and subtropical water from the Leeuwin Current System on shelf ecosystems is fundamental to understanding ecosystem dynamics and productivity. Determining the chemistry and nutrient makeup of each o f these water types can help manage ecosystem responses and predict levels of primary productivity during different times of the year, such as during peak summer upwelling events.

2 South Australia

Eyre Peninsula eastern GAB j / Spencer — 3 4 .5 Gulf r \ Gulf St. k S G O / t Vincent Lincoln Shelf

Upwelling Kangaroo Is.

Bonney Coast

Longitude (°E)

Figure 1.2. Upwelling region in South Australia between the eastern Great Australian Bight (GAB) and the Bonney Coast. Local currents include the South Australian Current (SAC), which is warm and relatively saline, Spencer Gulf Outflow (SGO), which is cold, saline and dense, and summer upwelling that is cool and fresh.

1.1 Motivation and scope of thesis

I approached this research topic wanting to understand the upwelling system between the eastern Great Australian Bight and the Bonney Coast. Many basic questions needed to be answered to understand the influence of upwelling on shelf ecosystems and commercial fisheries, such as: what were the upwelling characteristics and what was the hydrographic signature of upwelled water? What was the nutrient enrichment and did it vary between years? What water mass was being upwelled onto the shelf, and from what depth? Was the water mass from the subtropics (Leeuwin Current source), or from the subantarctic (Flinders Current source)? Physical and biological oceanography was currently being studied, yet no one was looking at the hydrochemistry of water on the shelf, what I considered to be properties linking the physical and biological systems together.

To define the characteristics of the upwelling system and put them into context, however, I first needed to understand shelf water masses and seasonality, to

3 understand the differences in upwelling to background conditions. But I realised I needed to go beyond that, and address the larger scale shelf and slope water masses for the southern margin, as so little was known about the water masses and their depth distribution on a regional scale. Previous studies of upwelling had focused only on the local scale shelf properties - a study was needed to link the upwelling system to the regional scale circulation to fully understand the origins of upwelled water and its significance to shelf ecosystems.

Central to understanding the regional scale water masses, was the Flinders Current. What were the characteristics of this current and how shallow was it? Was it shallow enough to source upwelled water off Kangaroo Island? What were the origins of the water masses transported by the current? And did this water come from the south or east of Australia? Did the water masses continue on the western margin of Australia as the Leeuwin Undercurrent flowing north?

I had previously used stable isotopes of seawater as a conservative proxy in water mass analyses. Based on the importance of source region determination and the occurrence of mixing between several different water masses in this study, I believed this proxy could be very useful when used in conjunction with conventional parameters of hydrochemistry, such as temperature, salinity, dissolved oxygen and the major nutrients nitrate, phosphate and silicate.

Based on the above motivation, I set out to answer all these questions in three stand- alone studies that are ultimately linked together. In addition to the water mass questions, I wanted to test if stable isotopes of seawater were useful proxies for such a study.

1.2 Thesis outline and key questions

This thesis comprises five chapters, three of which have been produced as stand- alone studies that report new data. These three studies are presented in Chapters 2, 3 and 4 and have been created for publication. As a result there may be repetition of background oceanography and methodologies between the chapters. The three studies are outlined below:

4 Chapter 2: Study #1: Shallow water masses and their connectivity along the southern margin of Australia

This study focuses on the regional-scale water masses in the top 1000 m for the southern margin of Australia. The main aim of this study is to determine the depth distribution and seasonal variation of water masses along the margin, to define the Flinders Current. Key questions include:

• What are the water masses in the top 1000 m? • What is their variation along the margin, and between seasons? • What are the source regions of these water masses? • Are they transported by the Leeuwin Current System or the Flinders Current? And hence what is the depth interface between these two current systems? • What is the hydrochemistry and depth distribution of the Flinders Current? Can it source upwelling in the Kangaroo Island upwelling region? • What is the connectivity between the southern Australian margin and the Subantarctic Zone south of Australia, via the Flinders Current? • What is the influence of the Tasman Outflow on water masses of the Flinders Cunent, hence defining connectivity between the eastern and southern margins of Australia? • How similar is the Flinders Current to the Leeuwin Undercurrent on the western margin of Australia? Is the Leeuwin Undercurrent a continuation of the Flinders Current? • Can stable isotopes help determine source regions and connectivity of water masses? • What are the stable isotope compositions of these water masses and how do they compare to the same water masses in other regions?

Chapter 3: Study #2: Water masses and their seasonal variation of the Lincoln Shelf, South Australia

This study focuses on shallow water masses of the Lincoln Shelf, in the Kangaroo Island upwelling region between the eastern Great Australian Bight and the Bonney Coast. The main aim of this study is to determine background conditions and

5 seasonal variation on the shelf, to allow for comparison with conditions during upwelling events. Key questions include:

• What water masses are present on the Lincoln Shelf, during summer, winter, and periods of upwelling? • What are the source regions or modes of formation of these water masses? • What is their spatial distribution during different times of year? • What is the water mass associated with upwelling? Does it come from the Flinders Current? • Can the path of upwelled water be identified on the shelf during upwelling events? • Can stable isotopes help identity mixing of water masses on the shelf, especially during upwelling events?

Chapter 4: Study #3; Upwelling characteristics and nutrient enrichment of the Kangaroo Island upwelling region, South Australia

This study focuses specifically on upwelling events in the Kangaroo Island upwelling region. The main aim of this study is to determine characteristics and nutrient enrichment during upwelling events, and compare these to background conditions on the shelf.

• What is the temperature and salinity of upwelled water on the shelf? • In more detail, what are the spatial, temporal and chemical characteristics of strong upwelling events? • What water mass is being upwelled and from what depth? How does this depth vary between upwelling events? • What are the nutrient characteristics of weak and strong upwelling events, and how do they compare to background conditions on the shelf?

Overarching these three studies is the application of stable isotopes of seawater to oceanography - are isotopes a useful method of identifying and tracing water masses in a predominantly evaporative region with significant latitudinal variation?

A discussion and conclusions chapter (Chapter 5) follows the three studies, linking each study together, highlighting findings and discussing implications of the work.

6 1.3 Methodology Overview

Water masses in the study area have been identified using both hydrochemistry and stable isotopes of seawater. Using these proxies together allows for greater understanding of water masses and mixing processes in regions of complex oceanography.

Hydrochemistry data

Temperature, salinity and the associated density characteristics have been used predominantly in the past to identify and trace water masses, as this data are readily obtained using electronic CTD profilers. In addition, density controls water mass formation and vertical stability. During mixing of water masses, however, temperature, and to a lesser extent salinity, changes and equilibrates quickly, making it difficult to identify initial water masses as they mix. Therefore in areas of significant mixing, such as shelf areas, temperature and salinity may not provide enough detail to understand water masses and mixing processes. Nutrient and dissolved oxygen data are also very useful in defining water masses, but do involve analysis of water and are more time consuming and less readily available. Dissolved oxygen is particularly useful in tracing high dissolved oxygen subantarctic water masses such as Subantarctic Mode Water from the subtropics towards the equator. Nutrients are useful proxies in areas of upwelling, but are often used up very quickly by organisms in the photic zone, and are therefore not good long-term tracers.

Stahle isotope data

Stable isotopes of seawater act as conservative tracers within a water mass because they are by far the major components of water. They are more conservative than temperature and salinity, and the stable isotope signal can be retained for longer when water masses mix. 6 O and 6‘H values refer to the ratios of 0 / O and “H/ H in seawater relative to the ratios in standard seawater, which is defined by the seawater standard Vienna-Standard Mean Ocean Water (V-SMOW). These isotope values are primarily controlled by surface evaporation, precipitation and freshwater input, which can vary substantially with latitude (Figure 1.3). The isotopic signature can then be modified by conservative mixing as the water mass is removed from the sea surface and transported away from its source region (e.g. Epstein and Mayeda,

7 1953; Lloyd, 1966; Gat et al., 1996; Frew et al., 2000).

^ x □ Ocean 0 / Water

Evaporative loss ^ from rivers reservoirs, lakes I m ore arid

Figure 1.3. Summary of how hydrologic processes, such as latitude, elevation, humidity, precipitation and evaporation, affect stable isotope compositions. Such hydrologic processes affect seawater in a similar way. MWL is the meteoric water line, which represents a global average relationship between oxygen and hydrogen isotopes in natural terrestrial waters. (Source: SAHRA, 2014.)

If the water mass is returned to the surface, such as in upwelling regions, isotope ratios can again be modified by ocean-atmospheric processes such as precipitation and evaporation, and by freshwater input from rivers and ice meltwater. For example, evaporation preferentially removes the lighter isotope from seawater, increasing isotope ratios, while precipitation and freshwater input add the lighter isotope to seawater, decreasing isotope ratios. Unlike salinity, the isotopic composition of freshwater and precipitation is also dependent on latitude. Precipitation at higher latitudes has lower isotopic ratios than those at the equator, based on the preferential “rain out” of the heavy isotope as air masses move poleward (Dansgaard, 1964).

Therefore, when a water mass forms at the sea surface, the latitude and surface conditions determine its isotopic composition. Once the water mass is removed from the sea surface, its isotopic composition is changed only by mixing. As a result, a water mass will have a distinct isotopic signature based on surface conditions in its source region, as well as transport pathways and mixing patterns as it moves away from the source region. This allows for stable isotopes to add another dimension to water mass studies, when used in conjunction with temperature and salinity.

8 While providing more conservative properties than temperature and salinity and greater insight into mixing processes, stable isotopes are expensive and time consuming to analyse. Modem techniques are allowing for greater throughput of samples and have improved the precision of hydrogen isotope analyses significantly. Automated and fast oxygen isotope analyses are also possible but precision is poorer than more time consuming methods, and this lower precision can pose problems in areas where isotope variation between water masses is small.

Apart from sparse global data from GEOSECS surveys in the 1970s, and two dozen or so studies since (see Schmidt et al., 1999), very few studies have utilised stable isotopes in seawater to understand oceanography, and very little isotope data exists for the Southern Ocean and around Australia. Additionally, oxygen isotope-salinity relationships have been the focus of most studies (Archambeau et al., 1998; Meredith et al., 1999; Khatiwala et al., 1999; Frew et al., 2000; Harwood et al., 2008; McConnell et al., 2009), while very few studies have used hydrogen isotopes. Both oxygen-hydrogen isotope relationships and hydrogen isotope-salinity relationships are still poorly understood with respect to latitudinal variation, differences with source regions, and mixing.

Isotope studies in shelf regions are usually in areas with large freshwater influx, where large salinity variations provide clear isotope-salinity relationships (e.g. Fairbanks, 1982; Khatiwala et al., 1999; Povinec et al., 2008). In comparison, subtropical shelf regions, where waters have been isotopically enriched by evaporation and where mixing mainly involves seawater only, have generally received less attention. Craig (1966) established clear positive relationships between salinity, oxygen and hydrogen isotopes using seawater samples from the Red Sea, where evaporation is the only process changing isotopic compositions. Gat et al. (1996) studied oxygen and hydrogen isotopes in the Mediterranean Sea and found high values of öl80 up to +2.2 per mil result from relatively dry and isotopically depleted continental air masses that led to high levels of evaporation and isotopic fractionation. However, increases in 0180 were not matched with increases in ö2H, due to the seasonal influx of 62H-depleted meteoric waters. Corlis et al. (2003) found non-linear relationships between salinity, öl80 and 62H in Spencer Gulf, Australia. They attributed these relationships to a combination of evaporation, internal mixing

9 and ocean exchange. Isotope-salinity relationships were governed by evaporation at the head of the gulf and by mixing processes between inflowing shelf water and outflowing high salinity water at the mouth of the gulf. Richardson et al. (2009) analysed isotope ratios from the eastern Great Australian Bight (GAB) and found large variation in ratios due to mixing of shelf waters, upwelled waters and highly evaporated waters formed locally at the head of the Bight. They found that outflow of the evaporated, high isotope waters mixed with slope waters to be then upwelled back onto the shelf. In the Cariaco Basin, seasonal variability of the oxygen isotope- salinity relationship was assessed by McConnell et al. (2009). They found distinct differences in this relationship between upwelling and non-upwelling seasons, with the strongest positive linear correlation found during periods of upwelling. These studies highlight the complex relationships governing seawater isotopes in evaporative regions, where evaporation is important but mixing and upwelling result in significant variability.

Stable isotope analyses

Water samples collected for stable isotope analysis in this study were measured at Queen’s Facility for Isotope Research, Queen’s University, in Kingston, Canada.

Oxygen isotopes were measured on a Finnigan GasBench II using a modified CO2-

H2O equilibration technique of Epstein and Mayeda (1953). 1.5 ml of sample was left to equilibrate for 48 hours with He + 3% CO2 gas of known composition at 25°C, and isotope ratios of the gas were then measured on a Finnigan DELTAplusXP stable isotope mass spectrometer. Hydrogen isotopes were measured on a Finnigan MAT 252 mass spectrometer connected to an automated Finnigan H-Device. Water samples were sent through a heated glass column with chrome metal to remove the oxygen and allow the H2 gas to enter the mass spectrometer. Measurements of :H/'H and lx0 /l60 ratios are presented in this thesis using standard “6” notation with respect to Vienna-Standard Mean Ocean Water (V-SMOW). The b value is defined as:

b (%o) - (R/Rv s mo w ~ 1) x 10’ where R is the isotope ratio 2H/’H or l80 /H,0. Repeated analysis of lab seawater standards showed reproducibility (la standard deviation) of ± 0.4 per mil (n = 194)

10 and ±0.13 per mil (n = 41) for 62H and Öl80 measurements, respectively.

1.4 References

Archambeau, A-S., C. Pierre, A. Poisson and B. Schauer (1998), Distributions of oxygen and carbon stable isotopes and CFC-12 in the water masses of the Southern Ocean at 30°E from South Africa to Antarctica: results of the CIVA1 cruise, Journal of Marine Systems, 17, 25-38

Cirano, M. and J. F. Middleton (2004), Aspects of the mean wintertime circulation along Australia’s southern shelves: Numerical studies, Journal o f Physical Oceanography, 34(3), 668-684.

Corlis, N. J., H. H. Veeh, J. C. Dighton and A. L. Herczeg (2003), Mixing and evaporation processes in an inverse estuary inferred from delta H-2 and delta 0-18, Continental Shelf Research, 2i(8), 835-846.

Craig, H. (1966), Isotopic composition and origin of the Rea Sea and Salton Sea geothermal brines, Science, 154(3156), 1544-1548.

Dansgaard, W. (\964), Stable isotopes in precipitation, Tellus, 16(4), 436-468.

Epstein, S., and T. Mayeda (1953), Variation of 0-18 content of waters from natural sources, Geochimica Cosmochimica Acta, 4(5), 213-224.

Fairbanks, R. G. (1982), The Origin of Continental Shelf and Slope Water in the

New York Bight and Gulf of Maine: Evidence from H21s0/H2160 Ratio Measurements, Journal of Geophysical Research, 87(C8), 5796-5808.

Frew, R. D., P. F. Dennis, K. J. Heywood, M. P. Meredith, and S. M. Boswell (2000), The oxygen isotope composition of water masses in the northern North Atlantic, Deep-Sea Research I, 47(12), 2265 - 2286.

Gat, J. R., A. Shemesh, E. Tziperman, A. Hecht, D. Georgopoulos, and O. Basturk (1996), The stable isotope composition of waters of the eastern Mediterranean Sea, Journal o f Geophysical Research, 101(C3), 6441-6451.

11 Harwood, A. J. P., P. F. Dennis, A. D. Marca, G. M. Pilling and R. S. Millner (2008), The oxygen isotope composition of water masses within the North Sea, Estuarine, Coastal and Shelf Science, 78, 353-359.

James, N. P., T. D. Boreen, Y. Bone and D. A. Feary (1994), Holocene carbonate sedimentation on the west Eucla Shelf, Great Australian Bight - A shaved shelf, Sedimentary Geology, 90(3-4), 161-177.

Khatiwala, S. P., R. O. Fairbanks and R. W. Houghton (1999), Freshwater sources to the coastal ocean off northeastern North America: Evidence from H21S0/H2I60, Journal o f Geophysical Research, 104( C8), 18241-18255.

Lewis, R. K. (1981), Seasonal upwelling along the southeastern coastline of South- Australia, Australian Journal of Marine and Freshwater Research, 32(6), 843- 854.

Lloyd, R. M. (1966), Oxygen isotope enrichment of sea water by evaporation, Geochim. Cosmochim. Acta, 20(8), 801-814.

McCartney, M. (1982), The subtropical recirculation of mode waters, Journal of Marine Research, 40 (Suppl.), 427-464.

McClatchie, S., J. F. Middleton and T. M. Ward (2006), Water mass analysis and alongshore variation in upwelling intensity in the eastern Great Australian Bight, Journal o f Geophysical Research-Oceans, 111( C8), article C08007.

McConnell, M. C., R. C. Thunell, L. Lorenzoni, Y. Astor, J. D. Wright and R. Fairbanks (2009), Seasonal variability in the salinity and oxygen isotopic composition of seawater from the Cariaco Basin, Venezuela: Implications for paleosalinity reconstructions, Geochemisty, Geophysics, Geosystems, 10(6), Q06019, 15pp.

McLeay, L. J., S. J. Sorokin, P. J. Rogers and T. M. Ward (2003), Benthic Protection Zone of the Great Australian Bight Marine Park: 1. Literature Review, 70 pp., South Australian Research and Development Institute (Aquatic Sciences), West Beach, South Australia. 12 Meredith, M. P., K. E. Grose, E. L. McDonagh, and K. J. Heywood (1999), Distribution of oxygen isotopes in the water masses of Drake Passage and the South Atlantic, Journal of Geophysical Research, 104( C9), 20,949-20,962.

Middleton, J. F. and M. Cirano (2002), A northern boundary current along Australia’s southern shelves: The Flinders Current, Journal o f Geophysical Research- Oceans, 107(C9), article 3129.

Middleton, J. F. and G. Platov (2003), The mean summertime circulation along Australia’s southern shelves: A numerical study, Journal of Physical Oceanography, 33(11), 2270-2287.

Middleton, J. F. and J. A. T. Bye (2007), A review of the shelf-slope circulation along Australia’s southern shelves: Cape Feeuwin to Portland, Progress in Oceanography, 75(1), 1-41.

Povinec, P. P., J. de Oliveira, E.S. Braga, J. -F. Comanducci, J. Gastaud, M. Groening, I. Fevy-Palomo, U. Morgenstern and Z. Top (2008), Isotopic, trace element and nutrient characterization of coastal waters from Ubatuba inner shelf area, south-eastern Brazil, Estuarine, Coastal and Shelf Science, 76, 522- 542.

Richardson, L. E., T. K. Kyser, N. P. James and Y. Bone (2009), Analysis of hydrographic and stable isotope data to determine water masses, circulation, and mixing in the eastern Great Australian Bight, Journal o f Geophysical Research, 144, C l0016, 14pp.

SAHRA (2014), Isotopes and Hydrology, Sustainability of semi-Arid Hydrology and Riparian Areas, retrieved January 22, 2014, from http://web.sahra.arizona.edu/programs/isotopes/oxygen.html

Schahinger, R. B. (1987), Structure of coastal upwelling events observed off the southeast coast of South-Australia during February 1983-April 1984, Australian Journal o f Marine and Freshwater Research, 38(4), 439-459.

Schmidt, G.A., G. R. Bigg and E. J. Rohling (1999), Global Seawater Oxygen-18

13 Database - v 1.21, http://data.giss.nasa.gov/ol8data/

Speich, S., B. Blanke, P. de Vries, S. Drijfhout, K. Döös, A. Ganachaud and R. Marsh (2002), Tasman leakage: a new route in the global ocean conveyor belt, Geophysical Research Letters, 29{ 10), 1416, doi: 10.1029/2001GL014586.

Van Ruth, P. D., G. G. Ganf and T. M. Ward (2010a), Hot-spots of primary productivity: An Alternative interpretation to Conventional upwelling models, Estuarine, Coastal and Shelf Science 90, 142-158.

Van Ruth, P. D., G. G. Ganf and T. M. Ward (2010b), The influence of mixing on primary productivity: A unique application of classical critical depth theory, Progress in Oceanography 85, 224-235.

Ward, T. M., L. J. McLeay, W. F. Dimmlich, P. J. Rogers, S. A. M. McClatchie, R. Matthews, J. Kämpf and P. D. van Ruth (2006), Pelagic ecology of a northern boundary current system: effects of upwelling on the production and distribution of sardine (Sardinops sagax), anchovy (Engraulis australis) and southern bluefin tuna (Thunnus maccoyii) in the Great Australian Bight, Fisheries Oceanography, 15(f), 191-207.

14 Chapter 2. Shallow water masses and their connectivity along the southern margin of Australia

Abstract

Four water masses are identified and described for the top 1000 meters water depth of the southern Australian continental slope, from Cape Leeuwin to Tasmania. Three are identified from previous literature on the southeast Indian Ocean: Subtropical Surface Water (STSW), Tasmanian Subantarctic Mode Water (TSAMW) and Tasmanian Intermediate Water (TIW), and one is newly identified and named: South Australian Basin Central Water (SABCW). Water masses are described using hydrographic and nutrient data as well as stable isotope values of seawater. STSW is transported east by the Leeuwin Current system and is modified by heating and evaporation along the subtropical continental shelf. SABCW is formed at the subtropical front within the South Australian Basin at ~40°S, TSAMW is formed within the Subantarctic Zone southwest of Tasmania, and TIW is formed from mixing of two different types of Antarctic Intermediate Water west of Tasmania. The Flinders Current then transports all three Southern Ocean water masses west along the continental slope. Stable isotope values show these water masses continue north along the Western Australian slope, identifying connectivity between the Flinders Current and Leeuwin Undercurrent. Water mass distributions both along the slope and with depth provide the first evidence of the depth distribution of the Flinders Current, and the interface between subantarctic water transported by it and subtropical water transported by the Leeuwin Current system. This interface is typically ~300 m deep during winter and ~250 m deep during summer, but can be as shallow as 150 m during summer in the Kangaroo Island upwelling region and off western Tasmania. This water mass distribution has implications for deep upwelling events off Kangaroo Island, which will supply characteristic Southern Ocean water to continental shelf ecosystems.

15 2.1. Introduction

The Flinders Current (FC; Bye, 1972, 1983) is a northern boundary current that flows west along the continental slope of southern Australia (Figure 2.1). Modelling studies have identified this westward flow and determined its core depth and seasonal variation (Middleton and Cirano, 2002; Middleton and Platov, 2003; Middleton and Bye, 2007). The FC has maximum transport within the permanent thermocline and extends from the surface to 2000 m or so. It increases in magnitude from 5 cm/s in the east to 20 cm/s in the west. Sverdrup transports are greatest in early summer; however, maximum current speeds are greater in winter (Middleton and Bye, 2007). The core of the FC should be weaker and shallower in summer, with maximum speeds of 5-10 cm/s at around 300-400 m depth, while maximum speeds reach 10-15 cm/s at around 600 m depth during winter (Middleton and Platov, 2003; Middleton and Cirano, 2004). Current meter observations of the FC, however, are scarce and present ambiguous data, which has been attributed to mesoscale eddy activity along the southern margin of Australia (Wood and Terray, 2005). The current has been observed in the west, with maximum speeds of 20 cm/s at 400-600 mwd (Cresswell and Peterson, 1993), but its occurrence in the east is still uncertain. Transport may be a result of general westward drift rather than a distinct westward current. In addition, the FC has been defined as westward flow, without regard to hydrographic properties. As a result, the hydrographic properties and depth distribution of the current are poorly understood. Data are needed to validate and augment the existing modelled results of the current, and clarify what water masses are being transported west along the southern margin of Australia.

The positive wind stress curl in the Southern Ocean drives an equatorward Sverdrup transport that creates the FC (Cirano and Middleton, 2004). This results in a weak anticyclonic gyre in the South Australian Basin, with the eastward Antarctic Circumpolar Current forming the southern limb and the westward FC forming the northern limb along the Australian continental slope (Schodlok and Tomczak, 1997; McCartney and Donohue, 2007; Figure 2.1). The FC therefore originates within the Subantarctic Zone, and is thought to transport Subantarctic Mode Water (SAMW) and Antarctic Intermediate Water (AAIW) north across the Subtropical Front (STF) and west along the Australian continental slope (McCartney and Donohue, 2007; Speich et al., 2002; Barker, 2004). The FC is also fed by the Tasman Outflow, a 16 remnant o f the East Australian Current that sinks and flows westward around the southern tip of Tasmania (Rintoul and Sokolov, 2001; Speich et al., 2002; Cirano and Middleton, 2004). Modelling of the Tasman Outflow by Speich et al. (2002) suggests that water originating in the Pacific entrained in the Tasman Outflow becomes an important component of the westward flow along the southern Australian margin. The model suggests that an intense westward flow of AAIW and SAMW south of Australia is present at both 145°E and 115°E.

AUSTRALIA Leeuwin Current

South Australian Current Spencer Gulf outflow

Subantarctic Zone West wind drift tsman Outflow Subantarctic Mode Water formation /

110°E 120°E 130°E 140°E 150°E Longitude (°E)

Figure 2.1. Regional circulation o f southern Australia. Surface currents - solid line, deeper currents - dashed line. STF = Subtropical Front. Question marks denote uncertainty with regional connectivity.

SAMW and AAIW are major water masses that form in the Southern Ocean. SAMW was first named by McCartney (1977), who identified a vertically homogeneous layer of temperature, called a thermostad, forming from deep winter convection just north o f the Subantarctic Front within the Southern Ocean. Such a thermostad, which he named Mode Water, is high in dissolved oxygen content and subducts into the ocean interior, playing an important role in ventilating the lower thermocline o f the southern hemisphere subtropical gyres (McCartney, 1982). SAMW forms within the Subantarctic Zone at several locations around the globe, and properties vary based on the latitude of the Subantarctic Front at that location (McCartney, 1977). South of Australia, the coldest and densest SAMW of the Indian Ocean forms southwest of

17 Tasmania (McCartney, 1977; Thompson and Edwards, 1981; Koch-Larrouy et al., 2010). Barker (2004) identified formation of Tasmanian SAMW (TSAMW) in a small area bounded by 140 to 145°E and 45 to 50°S. This type of SAMW is a result of air-sea interaction and deep winter convection within the region, as well as input of warm, saline subtropical water transported south by the East Australian Current. Rintoul and Bullister (1999) also suggest that in this region SAMW formation is influenced by northward Ekman drift of cold and fresh Antarctic Surface Water across the Subantarctic Front. TSAMW forms a thermostad and is transported northwest through the South Australian Basin with an average temperature and salinity of 9°C/34.65, and an average depth of 500 meters water depth (mwd) (Barker, 2004).

Tasmanian Intermediate Water (TIW) is a type of AAIW that is formed west of Tasmania from mixing of cold, fresh AAIW from the Antarctic Circumpolar Current south of Australia and wanner, more saline AAIW from the Tasman Sea (Barker, 2004). A small volume of AAIW detaches from the Antarctic Circumpolar Current south of Australia and moves north towards western Tasmania. This northward flowing component is cold and fresh ‘new’ AAIW (Sokolov and Rintoul, 2000). Offshoots of AAIW from the Antarctic Circumpolar Current also flow north to the east and west of New Zealand. These offshoots circulate and mix with lower latitude waters within the Tasman Sea and the Pacific subtropical gyre to become warmer, more saline and lower in dissolved oxygen (Rochford, 1960; Rochford, 1968; Sokolov and Rintoul, 2000). This ‘older’ AAIW is eventually incorporated into the East Australian Current and is transported south to Tasmania. The Tasman Outflow flows west around southern Tasmania and produces a warm, saline offshoot of AAIW into the South Australian Basin (Rintoul and Sokolov, 2001; Barker, 2004). TIW forms when this older more saline AAIW with a low salinity core of 34.4-34.47 mixes with the new, cold and fresh AAIW with a low salinity core of 34.35-34.4 coming directly from the Antarctic Circumpolar Current. TIW is identified as having intermediate properties between these two AAIW types and has a smooth T-S curve with few kinks. Once formed, TIW flows west across the South Australian Basin (Barker, 2004).

Above the FC, the eastward flowing Leeuwin Current system extends from the western margin of Australia to the west coast of Tasmania. This current system, 18 which includes the Leeuwin Current (LC), South Australian Current and Zeehan Current (Ridgway and Condie, 2004; Figure 2.1), has been relatively well studied in comparison to the FC, especially on the west coast of Australia. The LC (Cresswell and Golding, 1980) is an anomalous poleward eastern boundary current that flows south along the western Australian coast and east along the southern Australian shelf. It forms a strong (>150 cm/s at times), narrow (<100 km wide), oligotrophic surface current that flows south along the shelf break against the prevailing winds (Batteen et al., 2007). The LC becomes saltier, cooler and denser as it flows south and east due to subtropical additions, air-sea interactions and eddy mixing with Indian Ocean and Southern Ocean water (Cresswell and Griffin, 2004; Domingues et al., 2007). The LC is strongest from May to October and reaches a minimum in January (Holloway and Nye, 1985; Rochford, 1986). Its influence along the southern shelf is predominantly in winter, where it flows along the shelf edge as far as 135°E (Ridgway and Condie, 2004; Cirano and Middleton, 2004). The LC mixes with saline outflow from the Great Australian Bight (GAB) at this point, and continues to flow eastward as the South Australian Current. At Bass Strait, the flow continues south as the Zeehan Current (Ridgeway and Condie, 2004).

The LC has been well defined by hydrographic properties to the west and south of Western Australia, as a warm, low salinity flow of tropical water during winter and more saline flow of subtropical water during summer (Cresswell and Peterson, 1993; Domingues et al., 2007). The South Australian Current and Zeehan Current, on the other hand, have not been well defined hydrographically; however the South Australian Current is associated with high salinity outflow coming from the GAB. Surface waters along the southern margin have high temperature and salinity due to their subtropical origin, and have been generally referred to as subtropical water or Subtropical Surface Water (STSW), defined by James and Bone (2011) as having a temperature and salinity range of 10-22°C and 35.1-35.9, and an intermediate dissolved oxygen content of 220-245 pmol/L.

The shelf and slope ecosystems along the southern margin of Australia are still poorly understood. Recent studies have shown that there is high species richness and high levels of endemism within these ecosystems (Ward et al., 2006a; Currie and Sorokin, 201 la,b). These ecosystems are also economically important for Australia, supporting several major commercial fisheries, including sardine and southern 19 bluefin tuna (Ward et al., 2006b). Understanding both the depth range and water masses of the LC/FC system is critical to determine the relative influence of both Southern Ocean water masses and subtropical water masses on these ecologically and economically important ecosystems. In addition, the Tasman Outflow and FC together may be very important for transporting intermediate waters from the Pacific Ocean to the Indian Ocean, thereby offering an alternate route to the Indonesian Throughflow as part o f the global thermohaline circulation (Speich et al., 2002).

This paper aims to determine the water masses within the top 1000 mwd along the continental slope of southern Australia, by using hydrographic, nutrient and stable isotope data. Water mass distribution w ill then be used to determine the depth range and variation o f the LC/FC system based on the source region and path of these water masses around Australia. Such observational hydrographic data w ill also be useful in validating the current model results of the FC.

2.1.1 Stable Isotopes of Seawater

Stable isotopes of seawater act as conservative tracers within a water mass because they are the major components of water. öl80 and 62H values refer to the ratios of l80 /160 and :H/‘H in seawater relative to the ratios in standard seawater. These values are primarily controlled by conditions within the formation region of a water mass, such as surface evaporation, precipitation and freshwater input, which vary substantially with latitude. The isotopic signature can then be modified by conservative mixing as the water mass is removed from the sea surface and transported away from its source region (e.g. Epstein and Mayeda, 1953; Lloyd, 1966; Gat et al., 1996; Frew et al., 2000). Evaporation preferentially removes the lighter isotope from seawater, increasing isotope ratios, while precipitation and freshwater input add the lighter isotope to seawater, decreasing isotope ratios. Unlike salinity, the isotopic composition o f freshwater and precipitation is also dependent on latitude. Precipitation at higher latitudes has lower isotopic ratios than those at the equator, based on the preferential rain out of the heavy isotope as air masses move poleward (Dansgaard, 1964). When a water mass forms at the sea surface, the latitude and surface conditions determine its isotopic composition. Once the water mass is removed from the sea surface, its isotopic composition is changed only by mixing. As a result, a water mass w ill have a distinct isotopic signature based on

20 surface conditions in its source region, as well as transport pathways and mixing patterns as it moves away from the source region. Therefore stable isotopes, in conjunction with temperature and salinity, add another dimension to water mass studies.

Most isotope studies have focused on oxygen isotope-salinity relationships, especially in polar regions (Archambeau et al., 1998; Meredith et al., 1999; Khatiwala et al., 1999; Frew et al., 2000; Harwood et al., 2008; McConnell et al., 2009). As a result, oxygen-hydrogen isotope relationships and hydrogen isotope- salinity relationships are still poorly understood with respect to latitudinal variation, differences between source regions, and mixing in the ocean. In addition, stable isotope data are virtually non-existent around Australia. Stable isotope signatures for SAMW and AAIW have only been measured in a few studies from the Atlantic and eastern Pacific Oceans; identifying isotopic signatures of SAMW and AAIW for the Australian sector of the Indian Ocean is critical for comparison with these studies, and to understand differences in source regions, formation, and mixing pathways of these major water masses on a global scale.

2.2. Methods

This paper analyses hydrographic and stable isotope data collected from several RV Southern Surveyor voyages undertaken during 2010, from locations between the Tasman Sea and the Western Australian slope (Figure 2.2; see Appendix A). Voyages include SS2010 V01 in January-February to the south Tasman Sea, SS2010 V02 in March to the Subantarctic Zone southwest of Tasmania (46°S, 140- 142°E), SS2010 T01 in April along the southern and western Australian margins, SS2010 V06 in August on the continental shelf and slope west of Perth, and SS2010 T02 in August across the southern Australian margin. CTD profiles to 1000 mwd were taken and water samples were collected from Niskin bottles at twelve depths for nutrient and stable isotope analyses. This dataset is augmented with historic hydrographic data stored in the Commonwealth Scientific and Industrial Research Organisation (CSIRO) databases (CSIRO, 2012) and ARGO float data collected from 2004 to 2010.

21 All references to density p(S,T,p) throughout this chapter refer to potential density with a reference pressure of p = 0 m (p(S,T,0)). In-situ temperature rather than potential temperature was used to calculate potential density, therefore compression effects but not adiabatic effects were accounted for. However, for the data presented here, the errors relating to this are less than 0.3% (see Appendix B). For convenience, potential density is reported as sigma-t (ot) throughout this chapter, and is defined as:

ot = p(S,T,0)- 1000 kg/m3

The equation of state used in this chapter is UNESCO EOS-80 using practical salinity units. Pressure data is here reported as depth in meters, as 1 m is approximately equal to 1 decibar (within 1% error for the depth range 0-1200 m).

North West Cape

AUSTRALIA

Perth Esperance Albany Adelaide

Legend • SS2010_v01 (Feb) ★ SS2010_v02 (Mar) ■ SS2010J01 (Apr) ▲ S S 2010J02 (Aug) 4 SS2010_v06(Aug)

Longitude (°E)

Figure 2.2. CTD stations with nutrient and stable isotope samples from RV Southern Surveyor voyages during 2010.

Water samples were analysed for nitrate (NO3), phosphate (PO4) and silicate (SiCE) during voyages onboard the RV Southern Surveyor, and processed by CSIRO. Si*

(Si* = [SiCE] - [NO3]) is a proxy utilized in this study as a conservative tracer of 22 Southern Ocean water masses, in particular SAMW (Sarmiento et al., 2004; Bibby and Moore, 2011). Water samples for stable isotope analysis were measured for oxygen and hydrogen isotopic compositions at the Queen’s Facility for Isotope Research in Kingston, Canada. Oxygen isotopic compositions were measured using a modified CO2-FEO equilibration technique of Epstein and Mayeda (1953) on a Finnigan GasBench II inline with a Finnigan DELTAplusXP stable isotope mass spectrometer. Flydrogen isotopic compositions were measured on a Finnigan MAT 252 mass spectrometer connected to an automated Finnigan H-Device. Measurements of 2H/’H and lsO/l(10 ratios are presented in standard “6” notation with respect to Vienna-Standard Mean Ocean Water (V-SMOW). The b value is defined as:

5 (%o) = (R/R v -s mo w - 1) X 103 where R is the isotope ratio 2H/'H or '*0/,('0. Repeated analysis of lab seawater standards showed reproducibility (la standard deviation) of ± 0.4 per mil and ±0.13 per mil for 6:H and 6lxO measurements, respectively.

2.3. Results and Discussion

Four water masses have been identified in the top 1000 mwd along the continental slope of southern Australia (Table 2.1). Three water masses, Subtropical Surface Water, Tasmanian Subantarctic Mode Water and Tasmanian Intermediate Water, correspond to those identified in previous studies of the southeast Indian Ocean and south of Australia (e.g. McCartney, 1977; Rintoul and Bullister, 1999; Tomczak and Godfrey, 2003; Barker, 2004) and one, South Australian Basin Central Water, is here named based on its formation region within the South Australian Basin. Each water mass is defined by temperature, salinity, dissolved oxygen, density and nutrients. Stable isotope values and trends are then analysed for each water mass to provide additional information on source regions and mixing pathways.

2.3.1 Data Analysis

Two sets of data were analysed to determine water mass properties and their distribution along the southern margin of Australia. The first dataset of historic CTD,

23 Table 2.1. Properties of water masses along the southern margin. STSW - Subtropical Surface Water, SABCW - South Australian Basin Central Wati TSAMW - Tasmanian Subantarctic Mode Water, TIW - Tasmanian Intermediate Water hydrology and ARGO profile data was analysed using scatter plots as well as plotting in longitudinal section to determine depth distribution and variation of properties along the southern margin. The second dataset of hydrographic and stable isotope data from specific voyages undertaken in 2010 was analysed in more detail and compared the southern margin properties (voyages SS2010 T01 and SS2010 T02), to water properties along the west coast of Australia (voyages SS2010 T01 and SS2010V06), in the Tasman Sea (voyage SS2010 V01) and within the Subantarctic Zone (voyage SS2010V02).

To examine the distribution of water masses, scatter plots of temperature versus salinity are presented for four regions along the southern margin (Figure 2.3). The CTD and hydrographic data shown here were selected from the slope for water depths between 150 and 1200 mwd, for western Tasmania (southeast region, 41.5- 43.5°S/144-146°E), the upwelling region between Kangaroo Island and the Bonney Coast (east region, 136-142°E), the central Great Australian Bight (GAB; central region, 130-132°E) and adjacent to Albany south of Western Australia (west region, 116-118°E). The upper water depth of 150 m was chosen to include the shelf break while excluding highly variable properties of inner shelf areas.

(a) Summer months (November-April) (£>) Winter months (May-October)

May 1988 STSW

July-August SABCW T-S kink ZC water TSAMW

X Albany (west) X GAB (central) X Bonney Coast (east) X west Tasmania (southeast)

Salinity Salinity

Figure 2.3. Temperature versus salinity for a) summer months and b) winter months for four regions along the southern margin: Albany (west), central GAB (central), Bonney Coast (east) and western Tasmania (southeast). Grey lines represent potential density in sigma-t, with a reference pressure of 0 m.

25 In summary, the data show little seasonal or longitudinal variation in the water masses deeper than 200 m, especially waters with temperature and salinity below 12°C and 35.1. The general linearity here indicates that vertical mixing of these water masses has taken place. This lack o f seasonal variation below a variable surface layer is a characteristic of central water, such as Indian Central Water and West Pacific Central Water, which form by deep vertical mixing at the Subtropical Front (Tomczak, 1999). At depths above 200 m, the scatter during summer is due to latitudinal variation in heating and evaporation/precipitation, the decreasing influence of the Leeuwin Current from west to east, and upwelling events in the Kangaroo Island-Bonney Coast region.

To examine the longitudinal distribution o f the data, temperature, salinity, density and dissolved oxygen were plotted for both winter (May to October; Figure 2.4) and summer (November to April; Figure 2.5). Vertical arrows indicate the vertical extents of each water mass. In addition, three latitudinal sections (not shown) were examined at 120°E, 130°E and 140°E between 50°S and the continental slope of Australia, to determine formation regions o f water masses within the South Australian Basin. Salinity of the 130°E latitudinal section is shown in Figure 2.8b.

2.3.2 Subtropical Surface Water (STSW)

STSW is the shallowest water mass along the southern margin. It is identified by a maximum in temperature (>12°C) and salinity (>35.1), and a minimum in nutrients (Figure 2.3; Table 2.1). Dissolved oxygen values between 225 and 250 pmol/L are also high. STSW properties can be quite variable due to intense surface heating and evaporation during summer, saline, dense outflows from the GAB and Spencer G ulf during winter, and upwelling o f cold, low salinity SABCW onto the shelf in the upwelling region between Kangaroo Island and the Bonney Coast (see Chapter 3 of this thesis).

Origins and spatial distribution

STSW is formed locally within the surface mixed layer along Australia’s southern shelves, from high surface heating and evaporation between 34°S and 38°S. The water mass is also augmented by low salinity tropical water and high salinity South Indian Central Water (SICW), which are transported by the Leeuwin Current from

26 the west coast of Australia, around Cape Leeuwin and east along the southern margin (Cresswell and Peterson, 1993). Tropical water carried by the Leeuwin Current during both summer and winter is identified by low density south of Western Australia (Figures 2.4c, 2.5c). The density of this water increases from west to east as it mixes with locally formed STSW, as well as being modified by surface heating and evaporation during summer and cooling during winter.

The maximum depth of STSW varies both along the margin and seasonally, between 100 and 300 mwd (Figures 2.4, 2.5). The bottom surface is deeper during winter than during summer. It is also deeper west of the GAB, and shallows towards the east. In addition, a high salinity plume from the GAB or Spencer Gulf is present during winter between 138-140°E down to at least 300 mwd, which influences salinity and density contours down to -400 mwd (Figures 2.4b, 2.4c). During winter, the ZC transports STSW past the southern tip of Tasmania. STSW properties cool significantly but retain a relatively high salinity (35.1-35.4), as the South Australian Current and Zeehan Current transport this water southeast. Surface temperatures decrease from ~15°C along the Bonney Coast, to ~12.5°C off western Tasmania, resulting in denser water that plots to the right of the T-S curve in western Tasmanian profiles (Figure 2.3b). A deep winter mixed layer of -400 m can form, mixing this cold, saline water into SABCW as well as influencing TSAMW properties.

During summer, however, STSW distribution is limited to west of Bass Strait and north of 38°S. Surface waters on the west Tasmanian shelf do not represent the high temperature and salinity signature of subtropical waters and instead reflect more temperate conditions (Figure 2.5b). At ~40°S west of Tasmania the 35.1 isohaline of SABCW outcrops at the surface, although temperatures are warmer than 12°C due to summer heating. Therefore, surface waters off western Tasmania are significantly more saline in winter compared to summer due to the Zeehan Current influence during winter. The absence of southward flow of the Zeehan Current during summer is supported by drift bottle data published in the 1960s (Newell, 1961; Vaux and Olsen, 1961). More recent studies, however, have found the Zeehan Current to flow year-round (Lyne and Thresher, 1994; Cresswell, 2000). Based on the water mass analyses in this study, the Zeehan Current is not transporting STSW south during

27 summer.

TEMPERATURE

110°E 120°E 130°E 140°E 150°E SALINITY

1200 110°E

110"E 120‘ E 130°E 140°E I50*£ (d ) DISSOLVED OXYGEN o

200

400 € g. 600 Q 800

1000

1200 110°E 120°E 130°E Longitude

Figure 2.4. Averaged winter (May to October) a) temperature (°C), b) salinity, c) potential density reported as sigma-t (kg/m3) and d) dissolved oxygen (pmol/L) along the southern margin from 110°E to 147°E. A - STSW, A] - core o f Leeuwin Current, B - SABCW, C - TSAMW, D -T IW , D l - core ofT IW .

28 110°E 120°E 130°E 140°E 150°E

110"E 120°E 13CTE 140°E 150'E

110°E 120°E 130°E UO°E 150°E Longitude

Figure 2.5. Averaged summer (November to April) a) temperature (°C), b) salinity, c) potential density reported as sigma-t (kg/m’) and d) dissolved oxygen (pmol/L) along the southern margin from 110°E to 147°E. A - STSW, B - SABCW, C - TSAMW, D - TIW, D, - core of TIW.

29 NITRATE

UO°E SILICATE

120°E 130°E UO°E

Figure 2.6. Averaged yearly nutrient concentrations in longitudinal section along the southern margin, from Cape Leeuwin (ll5°E) to southwest Tasmania (147°E). a) nitrate (pmol/L), b) phosphate (pmol/L), c) silicate (pmol/L) and d) Si* (pmol/L), the relative abundance of silicate to nitrate. A - STSW, A1 - core of Leeuwin Current, B - SABCW, C -TSAMW, D-TIW.

30 N utrients

Nutrients are generally low in STSW, especially in the west. Nitrate and phosphate are typically less than 5 pmol/L and 0.5 pmol/L, respectively (Figures 2.6, 2.7). Western Tasmanian surface waters have relatively high nutrients; in the top 150 mwd nitrate is 4-5 pmol/L and phosphate is 0.4-0.5 pmol/L. In the top 150 mwd west of the GAB, however, nitrate is less than 1 pmol/L and phosphate is less than 0.2 pmol/L. Values increase sharply between 150 and 200 mwd to match values in Tasmanian surface waters.

Nitrate (umol/L)

Si* (umol/L)

Figure 2.7. Nutrient profiles by water mass for the five stations sampled during August 2010 on Southern Surveyor voyage SS2010T02, a) nitrate, b) phosphate, c) silicate, d) Si* (units are all pmol/L). STSW - Subtropical Surface Water, divided into western Tasmania (Tas.) and west o f GAB; SABCW - South Australian Basin Central Water; TSAMW - Tasmanian Subantarctic Mode Water; and TIW - Tasmanian Intermediate Water.

There is slightly higher silicate in tropical water carried by the Leeuwin Current compared to locally formed STSW. This produces a decreasing gradient in silicate values from >3 pmol/L at 120°E to <1 pmol/L at 140°E (Figure 2.6c). Silicate remains low but begins to increase with latitude along the west coast of Tasmania, with values between 1-2 pmol/L (Figure 2.6c). Low nitrate values throughout and higher silicate values in the west result in higher Si* values in the west (Figure 2.6d). The Leeuwin Current has a high silicate signature o ff western Australia (Thompson et al., 2011); silicate values greater than 2 pmol/L and Si* values greater than zero

31 reflect the influence of Leeuwin Current tropical water on STSW along the southern margin.

2.3.3 South Australian Basin Central Water (SABCW)

A new water mass with reduced variability below the surface mixed layer, in a density band between 26.65-26.8 kg/m’, is here named South Australian Basin Central Water (SABCW). It is present between the variable surface layer of STSW and the defined picnostad layer of TSAMW, and is remarkably similar in T-S plots along the southern margin at temperatures and salinities 10-12°C and 34.8-35.1, respectively (Figures 2.3, 2.4, 2.5). Nitrate and phosphate values increase with depth, and silicate values increase only slightly, leading to lower Si* with depth (Figure 2.6; Table 2.1).

Origins and spatial distribution

Based on historic data and ARGO profiles, the properties of SABCW along the southern margin match properties at or just north of the STF within the South Australian Basin. Previous studies have located the STF where the 12°C isotherm crosses the 150 m depth level (Nagata et al., 1988; James et al., 2002; Tomczak et al., 2004). Tomczak et al. (2004) used properties of 12°C, 35.1 and 26.7 kg/m' at 150 mwd to identify crossing of the STF at ~40°S within the South Australian Basin west of Tasmania. James et al. (2002) also found that the STF was characterised by the 12°C isotherm, and by a salinity range between 34.9 and 35.4 at 150 mwd within the South Australian Basin. South of Tasmania, Rintoul and Bullister (1999) identified the STF as a drop in temperature from 11.5° to 9°C and salinity from 35.0 to 34.8, which are properties that match those of SABCW.

The reduced variability below the surface mixed layer along the southern Australian margin can be defined as central water forming at the STF between 37 and 45°S south of Australia. Central water forms at the subtropical convergence during winter by subduction (Figure 2.8a). Subduction involves winter convection and formation of a deep mixed layer, followed by downward movement of water along isopycnals driven by Ekman pumping. This water is isolated from the ocean surface when surface waters warm during spring and summer, and is subducted into the ocean interior to form a homogeneous water mass at depth (Stommel, 1979; Tomczak and

32 Hao, 1989; Williams et al., 1995; Tomczak, 1999). Once formed, SABCW would be transported east by zonal flow at the STF (Schodlok and Tomczak, 1997), and then north and west along the Australian southern margin by the Flinders Current. A summer latitudinal section of salinity at 132°E (Figure 2.8b) also suggests that SABCW may subduct and flow directly north to the Australian continental slope. Once reaching the slope, the water mass would then be incorporated into the Flinders Current flowing west. Latitudinal sections at 120°E and 140°E (not shown) suggest SABCW is formed more in the east than in the west. The volume of water with SABCW properties is greatest in the 140°E section, is significant in the 130°E section and is minor in the 120°E section. No winter data have been collected to create corresponding latitudinal sections for winter. Between 40 and 45°S, a deep winter mixed layer off western Tasmania may also contribute water to SABCW, despite being north of the STF. This would result in STSW carried by the Zeehan Current being incorporated into SABCW, which would then flow back west towards the Indian Ocean as the FC. This has been postulated by Koch-Larrouy (2010) for SAMW within the FC, who used Lagrangian analysis to determine sources and pathways of SAMW within the Indian Ocean. They found that SAMW formed southwest of Tasmania included waters sourced from the Leeuwin Current, indicating a retroflection of Indian Ocean surface water back into the Indian Ocean.

Density sections show the increased spacing of density contours between 26.7 and 26.8 kg/m1 (Figures 2.4c, 2.5c), identifying SABCW as a layer of reduced density change with depth. In depth profiles it is still distinct from the strong picnostad of TSAMW, therefore the volume of SABCW would not be as significant as TSAMW. However, the spreading of density contours and the decreased density change with depth of SABCW may indicate a similar formation mechanism to TSAMW. In addition, Karstensen and Tomczak (1998) suggest that the high dissolved oxygen content of SAMW south of Australia is a result of formation by convective overturning and close proximity to the source region. The high values of dissolved oxygen within SABCW and TSAMW along the southern margin suggest that both water masses could have formed by convective overturning and subduction during winter south of Australia (Figure 2.8a).

Temperature and salinity profiles show depth variability along the margin, likely due

33 to seasonality, eddies, high salinity outflows and upwelling. The thickness and depth range o f SABCW decreases from east to west and the upper surface is shallower in summer compared to winter. The water mass is -150 m thick east o f the Great

Australian Bight and decreases to 100 m thick in the west during summer. The

12°C/35.1 surface is below the seasonal thermocline at -100 mwd o ff the west coast o f Tasmania, -150 mwd in the upwelling region between Kangaroo Island and the

Bonney Coast, and drops to -300 mwd west o f 134°E (Figure 2.5). During winter the top surface o f SABCW is -300 mwd along the margin, but is closer to 400 mwd at ~138°E which may be a result o f saline outflow from Spencer G u lf and the eastern

Great Australian Bight (Figure 2.4).

(a) winter schematic winter heat loss STF T T SAF

convection forming winter mixed layer

Ekman pumpir and transport i

isopycnal surfaces

isopycnal surfaces

North Latitude South

(b) summer data SÄC SALINITY

i STSW

35°S 40°S 45°S 50‘S Latitude

Figure 2.8. Winter schematic (a) and summer latitudinal salinity section (b) o f water masses south o f Australia, (a) shows the convergence region around the Subtropical Front (STF). SABCW forms by convective overturning and formation o f a deep mixed layer during winter, and subduction of this water along isopycnal surfaces via Ekman pumping, (b) shows salinity contours and water mass distribution along a latitudinal section at 132°E, between the Australian continental slope and 50°S. SAC - South Australian Current, STSW - Subtropical Surface Water, SABCW - South Australian Basin Central Water, TSAMW - Tasmanian Subantarctic Mode Water, TIW - Tasmanian Intermediate Water, STF - Subtropical Front, SAF - Subantarctic Front.

34 Nomenclature

SABCW has similar characteristics to Indian Central Water (ICW). However, based on the data here as well as previous water mass and modelling studies (see below), it is argued that this water is being formed at the STF within the South Australian Basin south of Australia. Central Water by definition spans the whole permanent thermocline (Tomczak, 1999), incorporating TSAMW and TIW; however as SAMW and AAIW are such well known water masses within the Southern hemisphere, SABCW is discussed as the water above TSAMW, being transported by the FC west along the southern Australian continental slope.

Indian Central Water (ICW) was first defined by Sverdrup et al. (1942) and since then has been identified by different names and T-S characteristics. T-S characteristics have ranged from 7 to 25°C and 34.6 to 35.8 (Warren et al., 1966; McCartney, 1977; Emery and Meincke, 1986; Tomczak and Large, 1989; Karstensen and Tomczak, 1998). The mode of formation was originally identified as convective overturning during late winter at 40-45°S within the southeast Indian Ocean (Sverdrup et al., 1942), and hence encompassed SAMW (McCartney, 1977). This definition has since been modified to the water of the permanent thermocline within the Indian Ocean (Tomczak and Godfrey, 2003), which includes the salinity maximum formed within the subtropics at ~30°S. This salinity maximum has become the distinct feature of South Indian Central Water (SICW; Fieux et al., 2005; Woo and Pattiaratchi, 2008), a water mass which is identified as a salinity maximum below less saline tropical water on the Western Australian shelf, with temperature >12°C and salinity >35.1. This high temperature and salinity association was inappropriate in this study, however. Woo and Pattiaratchi (2008) actually include the T-S properties of SABCW within SAMW on the west Australian slope, despite the lack of a picnostad at SABCW level. The salinity maximum of SICW corresponds to the salinity maximum of STSW along the southern margin of Australia, therefore a different water mass was needed to differentiate the lower salinity layer below STSW.

Based on literature reviewed, there is no evidence of eastward flow at the STF between the southern Indian Ocean and the South Australian Basin, supporting the contention here for a local formation region for SABCW. Stramma (1992) mapped

35 eastward flow at the STF within the southern Indian Ocean and named the South Indian Ocean Current. However, this current did not continue into the South Australian Basin south of Australia but instead moved north up the western Australian continental margin to eventually be included in the Indian Ocean subtropical gyre. Stramma (1992) stated that it is unlikely that subtropical water from the South Indian Ocean Current flows eastward south of Australia. Schodlok and Tomczak (1997), modelled circulation within the South Australian Basin using observations made during a World Ocean Circulation Experiment (WOCE) voyage. They found that the high degree of density compensation across the STF at 120°E resulted in minimal mass transport, whereas eastward transport of 3.5 Sv was identified at the STF at 132°E. Their conclusions matched those of Stramma (1992); that the South Indian Ocean Current does not flow from the Indian Ocean across the GAB. They comment that eastward flow at the STF south of Tasmania and in the Tasman Sea may arise from eastward flow developing around 132°E. These conclusions match hydrographic data analysed in this study. SABCW identified at the STF in N-S transects at 120°E, 132°E and 140°E show that the SABCW layer is less developed in the west than in the east. This suggests formation is predominantly between 132°E and 140°E.

Further distinction between the South Australian Basin and Indian Ocean formation regions for SABCW is presented by Karstensen and Tomczak (1998). Using CFC and oxygen data to determine water mass ages, they described ICW and SAMW as having similar T-S characteristics but different formation modes and formation regions, and hence described them as separate water masses. Oxygen-rich SAMW dominated the lower thermocline in the far southeast Indian Ocean and was formed by convective overturning south of Australia, while ICW was formed west of 55°E by subduction. SABCW has very similar oxygen values to TSAMW and higher values than ICW. This suggests a similar formation mode and formation region for both SABCW and TSAMW within the South Australian Basin.

Nutrients

Nitrate and phosphate concentrations begin to significantly increase with depth within SABCW, while silicate begins to slowly increase (Figure 2.7). Nutrient profiles vary little within the SABCW depth range from east to west along the

36 margin. In SS2010 T02 winter data, SABCW is between 250 and 400 mwd; nitrate increases from 7 to 14 pmol/L, phosphate from 0.55 to 1 pmol/L and silicate from 1.5 to 5 pmol/L, while Si* decreases from -5 to -10 pmol/L. Nutrient concentrations of SABCW along the southern margin match those at the STF south of Tasmania recorded by Rintoul and Bullister (1999). They state that nitrate and phosphate are relatively high at the STF, >8 pmol/L for nitrate and >0.8 pmol/L for phosphate, while silicate is generally low, <5 pmol/L.

2.3.4 Tasmanian Subantarctic Mode Water (TSAMW)

TSAMW is identified as a layer of relatively constant density between 26.8 and 26.9 kg/m ' along the southern Australian continental slope. It is centred between 400 and 650 mwd with a temperature and salinity range of 8-10°C and 34.5-34.8 (Figures 2.3, 2.4, 2.5). This layer also has a weak dissolved oxygen maximum with values as high as 270 pmol/L, but values are variable seasonally and from east to west along the margin. The water mass has relatively high nutrients and a minimum in Si* (Table 2.1).

Origins and spatial distribution

TSAMW is named following Barker’s (2004) definition of SAMW that forms within the South Australian Basin. TSAMW forms southwest of Tasmania with average properties of 8.5°C/34.5, but these are quickly modified to 9°C/34.65 as the water mass subducts and spreads north and west through the South Australian Basin and along the Australian southern margin (Barker, 2004). TSAMW is the coldest and densest SAMW of the Indian Ocean (McCartney, 1977; Thompson and Edwards, 1981) and has been named previously as South East Indian SAMW (Koch-Larrouy et al., 2010). Properties of TSAMW along the southern margin also match SAMW properties at the Subantarctic Front at ~140°E presented by Rintoul and Bullister (1999) and Rintoul and England (2002), and at 145-155°E presented by Thompson and Edwards (1981).

The TSAMW picnostad top surface is consistently at -410 mwd west of 130°E along the southern margin. At ~140°E, the top surface shallows to ~350 mwd during summer and deepens to close to 500 mwd during winter (Figures 2.4, 2.5). This is likely a result of raised isotherms and upwelling during summer and warm, saline

37 outflow during winter. The picnostad layer is thicker and better defined in the east, especially during summer. As the water mass flows west the bottom surface is gradually eroded and the picnostad layer becomes less well defined.

Nutrients

Nitrate and phosphate concentrations continue to increase with depth, while silicate remains low (Table 2.1; Figure 2.6). This results in a minimum in Si* within TSAMW (Figure 2.7d). Si* measures the relative abundance of silicate to nitrate, and values within the Subantarctic Zone are the lowest in the global ocean. Si* values less than -10 pmol/L have been used to trace SAMW through both the southern and northern hemisphere permanent thermoclines (Sarmiento et al., 2004). The Si* values of TSAMW along the southern margin range from -6.6 to -16.7 pmol/L, with an average of — 11.4 pmol/L.

2.3.5 Tasmanian Intermediate Water (TIW)

TIW is identified along the southern margin of Australian as a minimum in salinity below TSAMW. It has low temperature (4-8°C) and low salinity (<34.5), and lower dissolved oxygen values than the water masses above it (<225 pmol/L) (Table 2.1). The TIW low salinity core has a salinity minimum of 34.39, temperature of 4.5- 4.6°C and density of -27.24 kg/m'. The core decreases in depth from 1050 mwd at 140°E to 900 mwd at 110°E along the southern margin (Figure 2.4b). Nutrients are high, especially silicate, which increases significantly within the TIW depth range (Figures 2.6, 2.7).

Origins and spatial distribution

TIW is a type of Antarctic Intermediate Water (AAIW) that is formed west of Tasmania from mixing of cold, fresh AAIW from the Antarctic Circumpolar Current south of Australia (salinity minimum of 34.35-34.4) and warmer, more saline AAIW from the Tasman Sea (salinity minimum of 34.4-34.47). The Tasman Outflow flows west around southern Tasmania and transports this warmer, more saline AAIW into the South Australian Basin (Rintoul and Bullister, 1999), where it mixes with the colder, fresher AAIW variety (Barker, 2004). TIW is identified as having intermediate properties between these two AAIW types.

38 Temperature-salinity (T-S) plots show increased variability at AAIW level off the western Tasmanian continental slope, with core temperature ranging from 4.1 to 4.8°C and core salinity ranging from 34.37 to 34.44. This is in contrast to low variability at SABCW and TSAMW level off western Tasmania, as well as low variability at AAIW level for regions further west along the southern margin. This variation is attributed to varying influence of warm, saline AAIW and cold, fresh AAIW through time. Comparison of summer and winter profiles show more variation and higher salinity during winter, although both seasons have both types of AAIW. T-S plots with warm, saline AAIW show a distinct kink in the T-S profile at TSAMW level (Figure 2.3b) and this kink is also more obvious in winter profiles. A kink is also observed in the salinity-isotope relationship from SS2010 T02 winter voyage data, which will be discussed in the next section. Increased influence of warm, saline AAIW off western Tasmania during winter suggests a stronger Tasman Outflow during this time. This is not in agreement with other results which suggest that the Tasman Outflow is strongest during summer, when the EAC is strongest (Rintoul and Sokolov, 2001; Middleton and Cirano, 2005).

TIW is present in profiles at AAIW level west of 142°E. TIW along the southern margin has low variability and properties between warm, saline AAIW and cold, fresh AAIW. This indicates mixing of the two different AAIW types along the western Tasmanian continental slope as they are transported north and west along the southern margin by the FC. Another possible mixing region is in the open ocean west of Tasmania, and TIW would be transported north to reach the continental slope around 39°S/142°E, bypassing the western Tasmanian continental slope. Barker (2004) found TIW throughout the South Australian Basin, with salinity contours suggesting formation directly west of Tasmania between 140 and 145°E. Herraiz-Borreguero and Rintoul (2011) also found this mixed AAIW type in the region between 40-45°S/140-145°E, while cold, fresh AAIW was found south of 45°S and warm, saline AAIW was found east of Tasmania.

Nutrients

Within TIW, nitrate and phosphate concentrations are greater than 24 pmol/L and 1.5 pmol/L, respectively. Silicate increases significantly with depth to values as high as 70 pmol/L, which results in Si* values as great as 35 pmol/L (Table 2.1; Figure

39 2.7). Concentrations within the low salinity core of TIW are consistently 30-35 pmol/L for nitrate and 2-2.25 pmol/L for phosphate along the southern margin (Figure 2.6). Silicate, however, ranges from 25 to 50 pmol/L. As the silicate contours are so close together, this large range may simply be a result of natural variation within the depth range.

2.3.6 Stable Isotope Analysis

Stable isotope values from samples collected along the southern margin during voyages SS2010 T01 and SS2010 T02 show a strong linear relationship between ö'*0 and 62H, with an R2 value of 0.9 (Figure 2.9a). Isotope values range from -0.4 to +0.46%o for 6lsO and -1.4 to +4%o for 62H. Stable isotope values from samples within the Tasman Sea, the Subantarctic Zone and from the western margin of Australia show good overlap with southern margin values (Figure 2.9b), and any differences are within error of the isotope measurements. Therefore there are no regional differences in the ö'*0-ö2H relationship. Isotope-salinity relationships do show some differences, however (Figure 2.10), specifically from profiles within the Tasman Sea and off North West Cape, WA (see Figure 2.2 for locations). A linear relationship for both öl80-salinity and 62H-salinity is present for samples within the top 600 mwd, between STSW and TSAMW; the 6l*0-salinity regression equation is y = 0.43x - 15.07 (R2 = 0.86), while the 6:H-salinity regression equation is y = 2.58x - 88.65 (R2 = 0.94). These strong linear relationships suggest local formation and mixing between TSAMW, SABCW and STSW within the South Australian Basin. The Global Network of Isotopes in Precipitation (IAEA/WMO, 2006) shows minimal data from the southern Ocean; however 6lsO values between -10 and

-14%o and 02H values between -70 and -102%o have been collected from 45 to 70°S in the southeast Pacific and southwest Indian Oceans. In addition, even small amounts of meltwater transported from the Polar Zone north across the Subantarctic Front could significantly influence isotopic values of TSAMW. Such cross-frontal exchange has been suggested as important for TSAMW formation (Rintoul and England, 2002). Therefore, it is plausible that the high isotope-salinity end member represents evaporative conditions along the southern margin, and the low isotope- salinity end member represents a freshwater component that matches the latitude of

40 formation for TSAMW. It is hard to draw conclusions, however, with so few regional precipitation and seawater data.

A change in slope is apparent in the isotope-salinity relationship between TSAMW and TIW along the southern margin, with a steeper gradient and lower isotope value of the freshwater component. This highlights the remote southerly formation region of AAIW and suggests the influence of Antarctic Surface Water or sea ice meltwater with low isotopic composition during its formation. A change in slope may also highlight the formation of AAIW sources remote from the South Australian Basin. In addition, the relationship within TIW shows more variation than the relationship within the top 600 mwd, possibly due to the different AAIW source waters mixing to form TIW. The isotope-salinity relationship at AAIW level off North West Cape, WA, is markedly different from TIW, and this is due to the presence of Northwest Indian Intermediate Water (NIIW) with high salinity and low dissolved oxygen values. The source of NIIW is within the Red Sea (Rochford 1964; Woo and Pattiaratchi, 2008), and the water mass has retained its high salinity and isotope signature of the formation region. Mixing with TSAMW results in a negative isotope-salinity slope (Figure 2.10). TIW off Perth (light blue x in Figures 2.9b and 2.10) shows some influence of the southward flowing NIIW in isotope and salinity values.

Subtropical Surface Water

Stable isotope values of STSW are high (öl80 >0.05%o, 62H >2.0%o) due to excess surface evaporation over precipitation within the subtropical region. From SS2010 T02 winter voyage data in water depths between 1000-2000 m, surface values are highest (ölxO/ö2H ~0.3/3.0%o) in the west due to the influence of SICW, sourced from the Indian Ocean subtropical gyre, and lowest (6I80/62H ~0.1/2.0%o) off western Tasmania due to the influence of SABCW, sourced from the STF south of Australia. Surface values from the central GAB are mid-way between these two extremes, reflecting a gradual decrease in isotope values and SICW influence from west to east. Isotope data for the continental shelf proper exists only for the Kangaroo Island-Spencer Gulf region, with average shelf values 0.3-0.4%o and 3.0- 4.0%o for öl80 and 62H, respectively. For more detail, see chapter 3 of this thesis.

41 (a)

y=6.02x + 1.60 R2 =0.90

+ ++ i

4- Southern Margin -H- +

(b)

STSW

SABCW and STF surface

TSAMW

NIIW @ NWC + Southern Margin TIW off Perth x NIIWatNWC ■ NWC (rest) o STF-SW Tas • SAZ-SWTas * SAZ-SETas a SE Tasmania slope TIW core

6180 (%o)

Figure 2.9. ö ^ O -ö ^ relationship for a) southern margin samples and b) southern margin samples compared with samples from the western margin o f Australia, the STF and SAZ south o f Australia, and the Tasman Sea. Water masses: STSW = Subtropical Surface Water, SABCW = South Australian Basin Central Water, TSAMW = Tasmanian Subantarctic Mode Water, TIW = Tasmanian Intermediate Water, NIIW = Northwest Indian Intermediate Water. Also NWC = North West Cape, EAC = East Australian Current, STF = Subtropical Front, and SAZ = Subantarctic Zone.

42 (a)

S le w 200m @ NWC + A + n ttt- + + \ + T f -H-

TW 5m @ NWC a + ti. ^ + t T + +* Top 600m SM 2 ■ y=2.58x - 88.65 ^+ + R2 =0.94 I CM • * ¥ * t ‘o° lO t n . +

y=5.41x - 187.16 R'=0.59 JA

SAZ Tasman Sea profile gt ‘NIIW y=-16.67x + 575.87 R 2=0.83

Salinity

0.5

0.4

+k A* + + 0.3 STF sw Tasmania * surface + + SAZ Tasman Sea + + 0.2 surface t + STSW SAZ sw Tasmania ~ °'1 + d surface + O 0 03 Tasman Outflow water „ + Southern Margin (SM) SABCW TIW 0 « Perth V 1*-) x NIIW at NWC ■ NWC (rest) ^ + + TIW ------TSAMW o STF - SW Tas - 0.3 t w xO* • SAZ-SW Tas * + ^ SAZ-SE Tas * SE Tasmania slope - 0.4 — NIIW @ NWC

- 0.5 35 35.2 Salinity

Figure 2.10. Isotope-salinity relationships for all isotope samples, a) 62H-salinity and b) 6lxO-salinity. Water masses: STSW = Subtropical Surface Water, SABCW = South Australian Basin Central Water, TSAMW = Tasmanian Subantarctic Mode Water, TIW = Tasmanian Intermediate Water, NIIW = Northwest Indian Intermediate Water, TW = Tropical Water, SICW = South Indian Central Water. Also SM = Southern Margin, NWC = North West Cape, EAC = East Australian Current, STF = Subtropical Front, and SAZ = Subantarctic Zone.

South Australian Basin Central Water

Stable isotope values in SABCW are very similar along the southern margin and show much less variation compared to STSW values. öl80 values -0.13 to +0.08%o and 62H values +1 to +1.7%o indicated SABCW during the SS2010 T02 winter voyage from the five CTD profiles between Albany (south WA) and western

43 Tasmania (see Figure 2.2). The four eastern profiles, between Esperance and western Tasmania, showed SABCW in the 250 mwd samples. Therefore, during this voyage the SABCW top surface was as shallow as 250 mwd along the southern margin, which is 50 m shallower than the averaged winter depth of SABCW indicated by historic hydrographic data. This indicates mixing of SABCW and STSW at the interface that is not differentiated by temperature and salinity.

Isotope values of SABCW along the southern margin match values from samples collected at the STF during voyage SS2010_v02 in March 2010 (see Figures 2.2, 2.9b). Two stations were sampled during this voyage, one had a low salinity surface signature, suggesting the STF was just to the north of it, but both stations had similar T-S properties at 100 mwd, of 11-13°C and 35.0-35.1. Isotope values for these samples at 100 mwd were between -0.1 and +0.08%o for 6lsO and between 1.2 and 1.8%o for 62H (Figure 2.11), within the range of SABCW along the southern margin. Importantly, isotope values of SABCW along the southern margin matched values at the STF west of Tasmania more than values at the STF east of Tasmania within the Tasman Sea. The Tasman Sea station sampled during voyage SS2010_v01 in February 2010 had low salinity and isotope values in the top 40 mwd, likely from precipitation in the surface mixed layer, but isotope values at 100 mwd were higher than SABCW values, >0.1%o and >2.4%o for Ö18O and 62H, respectively. These values are similar to STSW along the southern margin and suggest the influence of subtropical water from the southward flowing East Australian Current.

Tasmanian Subantarctic Mode Water

As with SABCW, stable isotope values are consistent within TSAMW along the southern margin. öl80 values were -0.25 to -0.1%o and 62H values were -0.1 to +l.l%o within the picnostad of TSAMW (Table 2.1) during both autumn voyage SS2010 T01 and winter voyage SS2010 T02. The TSAMW top surface identified by stable isotope values was at -410 mwd in all profiles except the central GAB profile, where the top surface was slightly raised at 380 mwd. The bottom surface was between 650 and 700 mwd.

No published isotope values are available for SAMW, however Subantarctic Surface Water within Drake Passage had 6iX0 values between -0.15 and -0.1 %o (Meredith et

44 al., 1999), in good agreement with TSAMW values in this study. In addition, surface values south o f the STF from voyage SS2010_v02 were between -0.16 and -0.1 %o in the top 125 mwd, directly matching TSAMW values along the southern margin (Figure 2.11).

V -V.O++ *

. o Wh Tasman Sea h+i+-i • o+ + + SAZ profile tHH-+h -H-++o + \ ■tf * o + + ■ a o + -FF + * + * # + + + -1^++^ \ SICW max -fh- + @ NWC +A c^±H-+ -I ' - + £ £ 600 + & + + + Q) -H- + o -tx + Q -fc + A + + A + + -A + Southern Margin +x> •+ TIW off Perth + O •)- -H+ x NIIWatNWC + + Tasman Sea * + + slope profile jf- ■ NWC (rest) o STF-SWTas x x • + • SAZ-SWTas hjr, a SAZ-SETas a SE Tasmania slope

200 -o . 5 -0.4 -0.3 -0.2 -0.1 0 0.1 0.2 0.3 0.4 0.5

Ö18'30 (%o) 52H (% o )

Figure 2.11. Isotope depth profiles for all isotope samples, a) 6lxO versus depth and b) 6:H versus depth. TIW = Tasmanian Intermediate Water, N IIW = Northwest Indian Intermediate Water, NWC = North West Cape, STF = Subtropical Front, SAZ = Subantarctic Zone.

Tasmanian Intermediate Water

Stable isotope values ranged from -0.49 to -0.25%o for ö1!<0 and -1.4 to -0.1%o for

6: H within the TIW range during voyage SS2010 T02. Two groupings of values were identified, one preferentially between 650-800 mwd (ö180/ö 2H approx.

-0.28/-0.2%o) and one between 800-1000 mwd (6I80 /6 2H approx. —0.36/— 1 %o, best seen in Figure 2.9b). This may be a result of mixing between TSAMW and TIW at

600-800 mwd depths, and a strong TIW core between 800-1000 mwd. The distinction is also clear in the isotope-salinity relationship, as salinity is higher between 600-800 mwd (Figure 2.10). Differences in isotope values between saline AAIW o ff western Tasmania and TIW along the rest of the southern margin are within error of the isotopic measurements, but mixing of different types of AAIW may account for increased variability in isotope values and a lower R2 value compared to isotope values within SABCW and TSAMW (e.g. Figure 2.10a).

45 Previously published isotope values for AAIW are minimal and from areas remote to southern Australia. Measurements for AAIW within the Subantarctic Zone south of Africa (Archambeau et al., 1998) are spatially the closest to this study, and 6lxO values are within the same range as TIW, between -0.35 and -0.15%o, with salinity between 34.22 and 34.34. Salinity is higher south of Australia, likely due to older AAIW from the Subantarctic Zone south of Australia, and high salinity AAIW from the Tasman Sea. AAIW measurements from Drake Passage, east of South America have average 6lxO values of-0.2%o (Meredith et ah, 1999). Isotope values in both of these studies are within range or slightly higher than isotope values for TIW in this study. Lack of isotope data for the Indian Ocean and western Pacific sectors of the Southern Ocean makes it hard to compare isotope values along the southern margin of Australia, however, as there are still large spatial unknowns as well as a lack of knowledge on the amount of polar water influence on AAIW modification in these areas. Antarctic Surface Water measured by Meredith et ah (1999) in Drake Passage had ö180 values between -0.5 and -0.3%o, therefore lower isotope values of AAIW in this study may reflect AAIW modification within the Subantarctic Zone due to an exchange of water across the Subantarctic Front, as suggested by Rintoul and England (2002) for TSAMW formation. In addition, a latitudinal section at 140°E during summer (not shown) shows low salinity water (<34.4) outcropping at the sea surface south of 48°S, suggesting surface modification of AAIW in this region. Rintoul and Bullister (1999), however, suggest that AAIW is transported into the region from the west, rather than being locally formed or modified. Without greater spatial coverage of hydrographic and stable isotope data south of Australia, exact formation or modification of AAIW in this region remains unclear.

2.3.7 Regional Connectivity

Water masses, nutrient properties and stable isotope values can be used to identify and trace the major current systems south of Australia. The eastward flowing Leeuwin/South Australian Current system can be distinguished from the westward flowing Flinders Current by nitrate to silicate ratios and Si* values, two proxies that record the relative abundance of nitrate to silicate. Tropical water transported by the Leeuwin Current on the west coast of Australia has relatively high silicate values (Thompson et al., 2011). This silicate enrichment is present south of Western

46 Australia (Figure 2.6c), and silicate concentrations in surface waters decrease from west to east along the southern margin as the Leeuwin Current flows east. In this study, high Si* values (>0 pmol/L) are used as a signature of the eastward flowing Leeuwin/South Australian Current system (Figure 2.6d).

In contrast, the Southern Ocean is known to have low silicate concentrations in surface waters (Rintoul and Bullister, 1999; Hutchins et al., 2001), and that Southern Ocean water masses, especially SAMW, have the lowest Si* values in the global ocean (Sarmiento et al., 2004; Bibby and Moore, 2011). Both SABCW and TSAMW have very low Si* values (<-5 pmol/L) as a result of their source region south of Australia. Such low Si* values in profiles along the southern margin are used as a signature for the core of the Flinders Current (Figure 2.6d).

Comparisons of isotope values (Figures 2.9b, 2.10, 2.11) for profiles from the southern margin, southern Tasman Sea, the STF and Subantarctic Zone southwest of Tasmania, and off the west coast of Australia (see Figure 2.2) provide additional information about water mass connectivity within the region. Oxygen-hydrogen isotope and isotope-salinity relationships are linear and correlated in the top 600 mwd, between STSW, SABCW and TSAMW for the southern and western margins of Australia, and the STF within the South Australian Basin. Variation was seen within the Tasman Sea and at North West Cape on the western margin, however (Figures 2.9, 2.10, 2.11). This suggests continuity between the subantarctic and subtropical regions south and west of Australia. The Flinders Current can transport water from the STF, north and west along the southern margin, and continue as the Leeuwin Undercurrent off Western Australia. Connectivity with the Leeuwin Undercurrent is discussed in more detail in the next section.

Therefore, SABCW, TSAMW and TIW all form south of Australia and are transported north towards the Australian coastline and west along the slope by the Flinders Current (Figure 2.12), based on the characteristics and distribution of water masses, their formation region, and knowledge of the regional ocean circulation. The Flinders Current has only been defined by modelling studies and scarce and variable current meter data. It is important to determine both the depth range and water masses of the current to understand the influence of Southern Ocean water masses on shelf and slope ecosystems.

47 The shallowest water mass transported by the Flinders Current is SABCW, which can be as shallow as 200 mwd during summer. Previous studies have identified the core o f the Flinders Current at 400-600 mwd, within the depth range of TSAMW (Cresswell and Peterson, 1993; Middleton and Bye, 2007). Therefore the presence of SABCW at 200 mwd means the current is much shallower than previously thought. The current was found to be shallower during summer, in agreement with previous modelling studies (Middleton and Bye, 2007). Model results also suggest that the current would have greater Sverdrup transports during summer. Dissolved oxygen levels >250 pmol/L at 125°E, between 200 and 550 mwd, suggest greater transport o f Southern Ocean water masses during summer, compared to during winter when dissolved oxygen levels o f 250 pmol/L are restricted to east o f 137°E (Figures 2.4, 2.5).

West Leeuwin AUSTRALIA Current South Australian Australian _ _ Current Current STSW SICW

f/SICW Zeehan South Indian t Current Ocean Current ICW ^

Subtropical Front Tasman Outflow saline AAIW TIW / Subantarctic Zone formation TSAMW_ Antarctic Circumpolar Current formation fresh AAIW

110°E 120°E 130°E 140°E 150°E Longitude (°E)

Figure 2.12. Schematic o f currents and water masses south o f Australia. Water masses: TW - Tropical Water, SICW - South Indian Central Water, STSW - Subtropical Surface Water, SABCW - South Australian Basin Central Water, TSAMW - Tasmanian Subantarctic Mode Water, TIW - Tasmanian Intermediate Water, A A IW - Antarctic Intermediate Water, ICW - Indian Central Water.

This is the first evidence that deep upwelling o ff Kangaroo Island and the Bonney Coast at depths greater than 200 mwd can source Flinders Current water from the Southern Ocean, (see also Chapter 3 and Chapter 4 of this thesis) which has only been hypothesised by various authors (Ward et al., 2006b; Middleton and Bye, 2007; Richardson et al., 2009). Deep upwelling events can transport SABCW of high nitrate, phosphate and relatively low silicate onto the shelf. Knowledge of minor and

48 trace nutrients in Southern Ocean waters can be extrapolated to these shelf regions where no knowledge exists of minor and trace nutrient enrichment during upwelling events. This is important for determining the effect these nutrients could have on biological productivity.

Leeuwin Undercurrent connectivity

Hydrographic and stable isotope data collected on the Western Australian slope, off Perth (31-34°S; Figure 2.2) match well with data collected along the southern margin (hydrographic data not shown; isotope data presented in Figures 2.9, 2.10, 2.11), suggesting connectivity of the Flinders Current and Leeuwin Undercurrent to at least this latitude. North of this latitude, at North West Cape (22°S; Figure 2.2), the TSAMW layer is most distinct as a high dissolved oxygen layer between 400 and 500 mwd. 6I80/62H values are -0.2/0.7%o, within the range of TSAMW along the southern margin.

SABCW is also present above TSAMW in the south, but its depth range decreases significantly towards the north, likely due to erosion of the layer by high salinity SICW. This SICW is transported from the Indian Ocean subtropical gyre by the West Australian Current, joins the Leeuwin Undercurrent ~32°S (Andrews, 1977) and flows both north and south (Figure 2.12). At North West Cape, SABCW is not observed, instead a high salinity and stable isotope signal of SICW is present between 250-300 mwd, above TSAMW (Figure 2.1 lb).

Below TSAMW, TIW is present in the south, with identical temperature, salinity and blsO characteristics to southern margin TIW. 62H values are slightly higher, and dissolved oxygen content is slightly lower, due to the influence of NIIW flowing south. At North West Cape, no TIW is observed. Instead NIIW is present below TSAMW as a high salinity, low dissolved oxygen water mass. This water mass is clearly seen in the isotope-salinity relationships (Figure 2.10). Historic hydrographic data and data from voyage SS2010 T01 suggest that north of 25°S, NIIW with salinity >34.5 and dissolved oxygen <170 pmol/L sits below TSAMW at -700 mwd. South of 25°S, NIIW deepens as it flows south below TIW. This is in agreement with water mass analyses between Cape Leeuwin and North West Cape by Woo and Pattiaratchi (2008), who found NIIW to flow south below the northward flowing AAIW. 49 The Leeuwin Undercurrent therefore appears to be an extension of the Flinders Current at the core depth of 400-500 mwd, transporting TSAMW to North West Cape. However, addition of SICW above the core, and southerly flow of NI1W below the core, results in the disappearance of both SABCW and T1W as the current flows north. While the Flinders Current transports water west at AAIW level, the Leeuwin Undercurrent competes with southward flow and mixing of NIIW, likely restricting the majority of flow to TSAMW level and shallower.

Tasman Sea influence

The presence of a warm, saline version of AAIW from the Tasman Sea on the western Tasmanian continental slope provides further evidence of the Tasman Outflow south of Tasmania. This westward outflow has been identified in only a few hydrographic studies (Rintoul and Bullister, 1999; Rintoul and Sokolov, 2001; Barker, 2004) on much larger spatial scales. This study shows that AAIW from the Tasman Sea has a significant influence on the western Tasmanian continental slope, and results in high variability at the AAIW depth level. Formation and transport of TIW along the southern margin provides evidence of intermediate water mass transport between the Pacific and Indian Oceans south of Australia, which is an alternative route to the Indonesian Throughflow.

Within the SAMW depth level, however, stable isotope values from eastern Tasmania do not match values from western Tasmania, suggesting the core of the Tasman Outflow is at AAIW level rather than shallower. The stable isotope profile within the Subantarctic Zone from the Tasman Sea matches southern margin profiles well in the 8,sO-salinity relationship but has lower 02H in the 82H-salinity relationship (Figure 2.10). The difference is also clear in the 82H depth profile (Figure 2.11b), with 82H values ~1.2%o lower between 300 and 700 mwd in the Tasman Sea Subantarctic Zone profile compared to other profiles. Values match again at AAIW level. The southeast Tasmania slope profile is also clearly different in the isotope-depth profiles for both öl80 and 62H. 82H values are ~l%o higher and öl80 values are 0.1-0.2%o higher at the equivalent depth in the southeast Tasmania slope profile. This slope profile does show a similar öl80-ö2H relationship to southern margin profiles (Figure 2.9b), but depth profiles show a shift of 200-300 m deeper for the equivalent water masses. Both Tasman Sea profiles are therefore

50 different to southern margin profiles, suggesting this region is not a significant source region of the Flinders Current at depth levels between 300 and 700 mwd.

Hydrographic properties analysed in this study are also not able to distinguish a different SAMW coming from the Tasman Sea. This is in agreement with hydrographic properties along the WOCE profile from Tasmania to Antarctica, in which the westward flow is strongest at the salinity minimum level (Rintoul and Sokolov, 2001). CTD profiles and geostrophic calculations done by Rintoul and Bullister (1999) and Rintoul and Sokolov (2001), however, suggest westward flow through most of the water column. In this study, depth distribution of water masses and stable isotope values do not suggest input of SAMW from the Tasman Sea to the southern Australian margin, however this is based on only two stable isotope profiles. For future research, increased spatial extent and spatial resolution of stable isotope data within the Tasman Outflow region may resolve this issue.

2.4. Conclusions

This study describes the water mass characteristics of the top 1000 meters water depth (mwd) along the southern continental margin of Australia, between 115°E and 145°E. Four water masses are transported along the margin by two current systems, the eastward flowing Leeuwin Current at the surface and the westward flowing Flinders Current with a core at 400-600 mwd (Figure 2.12). The interface between the two currents is between 200 and 250 mwd, where subtropical water overlies Southern Ocean water that has formed at the Subtropical Front south of Australia.

Three water masses correspond to those identified in previous water mass studies in the region, Subtropical Surface Water, Tasmanian Subantarctic Mode Water and Tasmanian Antarctic Intermediate Water, and one water mass is newly identified and named, South Australian Basin Central Water. While previous studies have focused on water mass formation in the Indian Ocean and within the Subantarctic Zone south of Australia, none have focused on the shelf and slope of the southern Australian continental margin. Hydrography and chemistry of the water in this extensive region is important for slope habitats as well as shelf ecosystems in upwelling areas such as Kangaroo Island and the Bonney Coast.

51 By increasing depth, these water masses are:

1) Subtropical Surface Water (STSW), with high temperature, high salinity, high isotope values, low nutrients and high variability; formed locally from surface heating and evaporation in the top 200 mwd; it is also augmented by tropical water and South Indian Central Water transported by the Leeuwin Current from the west coast of Australia; STSW is transported east by the Leeuwin Current, South Australian Current and Zeehan Current;

2) South Australian Basin Central Water (SABCW) within the upper permanent thermocline, with lower temperature, salinity and isotope values, high dissolved oxygen, increasing nutrients and lower variability; formed at the surface at the Subtropical Front (~40-45°S) south of Australia and present at -200-400 mwd along the southern margin; it is shallower in summer, especially in upwelling regions, and deeper in winter, and is transported west by the Flinders Current;

3) Tasmanian Subantarctic Mode Water (TSAMW), a thick layer of relatively constant density between 26.8 and 26.9 kg/nT with high dissolved oxygen and low Si* values; formed at the surface within the Subantarctic Zone southwest of Tasmania and present at -400-650 mwd along the southern margin; it is transported west by the Flinders Current and has an important role in ventilating the permanent thermocline of the Indian Ocean subtropical gyre;

4) Tasmanian Intermediate Water (TIW), with a salinity minimum, low temperature, dissolved oxygen and isotope values, and high nutrients; it has a low salinity core between 900 and 1000 mwd and is transported west by the Flinders Current; TIW along the southern margin is a mix of newly formed, cold, fresh AAIW formed south of the Subantarctic Front south of Australia and transported north, and older, more saline AAIW from the Tasman Sea transported around the southern tip of Tasmania by the Tasman Outflow.

Nitrate and phosphate concentrations along the continental margin increase with increasing depth. Silicate, however, is very low in the top 600 m, resulting in increasing nitrate to silicate ratios and decreasing Si* values with depth. Si*

52 measures the relative abundance of silicate to nitrate, and values in the Subantarctic Zone are the lowest in the global ocean. Both SABCW and TSAMW have very low Si* values (<-5 pmol/L) due to their source region south of Australia, and these low Si* values are used as a signature for the core of the Flinders Current. In contrast, tropical water transported by the Leeuwin Current south of Western Australia has relatively high surface water silicate values, and as a result silicate values in surface waters decrease from west to east. High Si* values (>0 pmol/L) are used as signature of the eastward flowing Leeuwin/South Australian Current system.

Stable isotope values of seawater decrease with increasing depth, with STSW having highest isotope values and TIW having lowest isotope values based on the latitude of their formation at the sea surface. A linear salinity-isotope relationship between STSW, SABCW and TSAMW is a result of their similar formation region and mixing patterns south of Australia. In contrast, the two AAIW types that mix to create TIW were formed south of 60°S in regions remote to southern Australia, and a change in the salinity-isotope relationship is evident between TSAMW and TIW as a result. In addition, isotope values at the STF within the South Australian Basin and west of Perth match well with isotope values along the southern margin, supporting the movement of water masses from the Subantarctic Zone, west along the southern margin and north along the western Australian margin by the Flinders Current and Leeuwin Undercurrent. In contrast, differences are seen in isotope values within the Tasman Sea and at North West Cape, Western Australia, due to different water mass influences external to southern Australia.

Based on the characteristics and distribution of water masses, their formation region, and the regional ocean circulation, SABCW, TSAMW and TIW all form south of Australia and are transported north towards the Australian coastline and west along the slope by the Flinders Current (Figure 2.12). It is important to determine both the depth range and water masses of the Flinders Current to understand the influence of Southern Ocean water masses on shelf and slope ecosystems along the southern margin. The Flinders Current transports SABCW that is found as shallow as 200 mwd on the slope during summer, which makes the current much shallower than previously thought. Deep upwelling off Kangaroo Island and the Bonney Coast transports SABCW of high nitrate, phosphate and relatively low silicate onto the shelf. Knowledge of minor and trace nutrients in Southern Ocean waters can be 53 extrapolated to these shelf regions where no knowledge exists of minor and trace nutrient enrichment during upwelling events, and the effect these nutrients have on biological productivity.

Acknowledgements. This research was supported by data collected through the Southern Australian Integrated Marine Observing System (SAIMOS). We thank the captain and crew of the R. V. Ngerin during numerous research voyages through all kinds of weather. We also thank the CSIRO Marine and Atmospheric Research division and captain and crew of the R.V. Southern Surveyor for their help and support during research voyages SS2010_v01, SS2010_v02, SS2010_v06, SS2010_t01 and SS2010_t02. Stable isotope analyses were done with the assistance of Kerry Klassen and April Vuletich in the Queen’s Facility for Isotope Research. Support of these analyses was from grants from the Natural Science and Engineering Research Discovery program, the Canadian Foundation for Innovation and the Ontario Innovation Foundation.

54 2.5. References

Andrews, J. C. (1977), Eddy structure and the West Australian Current, Deep-Sea Research, 24, 1133-1148.

Archambeau, A-S., C. Pierre, A. Poisson and B. Schauer (1998), Distributions of oxygen and carbon stable isotopes and CFC-12 in the water masses of the Southern Ocean at 30°E from South Africa to Antarctica: results of the CIVA1 cruise, Journal o f Marine Systems, 17, 25-38

Barker, P. M. (2004), The circulation and formation of water masses south of Australia and the inter-annual wind variability along the southern Australian coast, PhD thesis, University of Melbourne, Victoria, Australia, 351pp.

Batteen, M. L., R. A. Kennedy and H. A. Miller (2007), A process-oriented numerical study of currents, eddies and meanders in the Leeuwin Current System, Deep-Sea Research Part II-Topical Studies in Oceanography, 54(8- 10), 859-883.

Bibby, T. S. and Moore, C. M. (2011), Silicatemitrate ratios of upwelled waters control the phytoplankton community sustained by mesoscale eddies in sub- tropical North Atlantic and Pacific, Biogeosciences, 8, 657-666.

Bye, J. A. T. (1972), Oceanic circulation South of Australia, in Antarctic Oceanology II: The Australian-New Zealand Sector, Antarctic Research Series, vol. 19, edited by D. E. Hayes, pp. 95-100, AGU, Washington D. C.

Bye, J.A.T. (1983), Physical oceanography. In: Tyler, M.J., Twidale, C.R., Ling, J.K. and Holmes, J.W. (Eds.), Natural History of the South East, Royal Society of South Australia.

Cirano, M. and J. F. Middleton (2004), Aspects of the mean wintertime circulation along Australia’s southern shelves: Numerical studies, Journal of Physical Oceanography, 34(3), 668-684.

Cresswell, G. (2000), Currents of the continental shelf and upper slope of Tasmania,

55 Papers and Proceedings of the Royal Society of Tasmania, 133(3), 21-30.

Cresswell, G. R. and T. J. Golding (1980), Observations of a south-flowing current system in the southeastern Indian-Ocean, Deep-Sea Research Part I- Oceanographic Research Papers, 27(6), 449-466.

Cresswell, G. R. and J. L. Peterson (1993), The Leeuwin Current south of Western Australia, Australian Journal o f Marine and Freshwater Research, 44, 285- 303.

Cresswell, G. R. and D. A. Griffin (2004), The Leeuwin Current, eddies and sub- Antarctic waters off south-western Australia, Marine and Freshwater Research, 55(3), 267-276.

CSIRO (2012), Commonwealth Scientific and Industrial Research Organisation. Data provided by CSIRO and reproduced with the permission of CSIRO, accessed through the Data Centre: http://www.marine.csiro.au/datacentre/

Currie, D. R. and S. J. Sorokin (201 la), A comparison of megafaunal biodiversity in two contrasting submarine canyons on Australia’s southern continental margin, SARDI Research Report Series No. 519, South Australian Research and Development Institute (Aquatic Sciences), Adelaide, 49pp.

Currie, D. R. and S. J. Sorokin (201 lb), A preliminary assessment of the deepwater benthic communities of the Great Australian Bight Marine Park, SARDI Research Report Series No. 592, South Australian Research and Development Institute (Aquatic Sciences), Adelaide, 61pp.

Dansgaard, W. (1964), Stable isotopes in precipitation, Tellus, 16(4), 436-468.

Domingues, C. M., M. E. Maltrud, S. E. Wijffels, J. A. Church and M. Tomczak (2007), Simulated Lagrangian pathways between the Leeuwin Current Systsm and the upper-ocean circulation of the southeast Indian Ocean, Deep-Sea Research Part II-Topical Studies in Oceanography, 54(8-10), 797-817.

Emery, W. J. and J. Meincke (1986), Global water masses: Summary and review,

56 Oceanol. Acta, 9, 383-391.

Epstein, S., and T. Mayeda (1953), Variation of 0-18 content of waters from natural sources, Geochimica Cosmochimica Acta, 4(5), 213-224.

Fieux M., R. Molcard and R. Morrow (2005), Water properties and transport of the Leeuwin Current and Eddies off Western Australia, Deep-Sea Research I, 52, 1617-1635.

Frew, R. D., P. F. Dennis, K. J. Heywood, M. P. Meredith, and S. M. Boswell (2000), The oxygen isotope composition of water masses in the northern North Atlantic, Deep-Sea Research I, 47(12), 2265 - 2286.

Gat, J. R., A. Shemesh, E. Tziperman, A. Hecht, D. Georgopoulos, and O. Basturk (1996), The stable isotope composition of waters of the eastern Mediterranean Sea, Journal of Geophysical Research, 101( C3), 6441-6451.

Harwood, A. J. P., P. F. Dennis, A. D. Marca, G. M. Pilling and R. S. Millner (2008), The oxygen isotope composition of water masses within the North Sea, Estuarine, Coastal and Shelf Science, 78, 353-359.

Herraiz-Borreguero, L. and S. Rintoul (2011), Regional circulation and its impact on upper ocean variability south of Tasmania, Deep-Sea Research II, 58, 2071- 2081.

Holloway, P. E. and Nye, H. C. (1985), Leeuwin Current and Wind Distributions on the Southern Part of the Australian North West Shelf between January 1982 and July 1983, Australian Journal o f Marine and Freshwater Research, 36, 123-37.

Hutchins, D. A., P. N. Sedwick, G. R. DiTullio, P. W. Boyd, B. Queguiner, F. B. Griffiths and C. Crossley (2001), Control of phytoplankton growth by iron and silicic acid availability in the subantarctic Southern Ocean: Experimental results from the SAZ Project, Journal o f Geophysical Research, 106(C l2), 31,559-31,572.

57 IAEA/WMO (2006), Global Network of Isotopes in Precipitation, the GNIP Database, assessable at: http://www.iaea.org/water

James C., M. Tomczak, I. Helmond and L. Pender (2002), Summer and winter surveys of the Subtropical Front of the south eastern Indian Ocean 1997-1998, Journal o f Marine Systems, 37, 129-149.

James, N. P. and Y. Bone (2011), Neritic Carbonate Sediments in a Temperate Realm, Southern Australia, Springer, Berlin, 254pp.

Karstensen, J. and M. Tomczak (1998), Age determination of mixed water masses using CFC and oxygen data, Journal o f Geophysical Research, 103(C9), 18,599-18,609.

Khatiwala, S. P., R. O. Fairbanks and R. W. Houghton (1999), Freshwater sources to the coastal ocean off northeastern North America: Evidence from H2I80/H2I(,0, Journal o f Geophysical Research, I04( C8), 18241-18255.

Koch-Farrouy, A., R. Morrow, T. Penduff and M. Juza (2010), Origin and mechanism of Subantarctic Mode Water formation and transformation in the Southern Indian Ocean, Ocean Dynamics, 60, 563-583.

Floyd, R. M. (1966), Oxygen isotope enrichment of sea water by evaporation, Geochim. Cosmochim. Acta, 30(8), 801-814.

Fyne, V. D. and R. E. Thresher (1994), Dispersal and advection of Macruronus novaezealandiae (Gadiformes: Merlucciidae) larvaep off Tasmania: Simulation of the effects of physical forcing on larval distribution. In Sammarco, P. W. and Heron, M. L. (Eds.) The Bio-physics of Marine Larval Dispersal, Coastal and Estuarine Studies, 45, chapter 6, 109-136. American Geophysical Union, Washington, D.C.

McCartney, M. S. (1977), Subantarctic Mode Water. In: A Voyage of Discovery, M. Angel, editor. Supplement to Deep-Sea Research, 24, 103-119.

McCartney, M. (1982), The subtropical recirculation of mode waters, Journal of

58 Marine Research, 40 (Suppl.), 427-464.

McCartney, M. S. and K. A. Donohue (2007), A deep cyclonic gyre in the Australian-Antarctic Basin, Progress in Oceanography, 75(4), 675-750.

McConnell, M. C., R. C. Thunell, L. Lorenzoni, Y. Astor, J. D. Wright and R. Fairbanks (2009), Seasonal variability in the salinity and oxygen isotopic composition of seawater from the Cariaco Basin, Venezuela: Implications for paleosalinity reconstructions, Geochemisty, Geophysics, Geosystems, 10(6), Q06019, 15pp.

Meredith, M. R, K. E. Grose, E. L. McDonagh, and K. J. Heywood (1999), Distribution of oxygen isotopes in the water masses of Drake Passage and the South Atlantic, Journal o f Geophysical Research, 104(C9), 20,949-20,962.

Middleton, J. F. and M. Cirano (2002), A northern boundary current along Australia’s southern shelves: The Flinders Current, Journal o f Geophysical Research- Oceans, 107(C9), article 3129.

Middleton, J. F. and G. Platov (2003), The mean summertime circulation along Australia’s southern shelves: A numerical study, Journal o f Physical Oceanography, 35(11), 2270-2287.

Middleton J. F. and M. Cirano (2005), Wintertime circulation off southeast Australia: Strong forcing by the East Australian Current, Journal of Geophysical Research, 770(C12O12), doi:10.1029/2004JC002855.

Middleton, J. F. and J. A. T. Bye (2007), A review of the shelf-slope circulation along Australia’s southern shelves: Cape Leeuwin to Portland, Progress in Oceanography, 75(1), 1-41.

Nagata Y., Y. Michida and Y. Umimura (1988), Variation of positions and structures of the oceanic fronts in the Indian Ocean sector of the Southern Ocean in the period from 1965 to 1987. In Sahrhage D. (Ed), Antarctic ocean and resources variability, Springer, Berlin Fleidelberg New York, pp 92-98.

59 Newell, B. S. (1961), Hydrology of south-east Australian waters: Bass Strait and New South Wales tuna fishing area, CSIRO Div. Fish. Oceanogr. Tech. Pap. 10, 22pp.

Richardson, L. E., T. K. Kyser, N. P. James and Y. Bone (2009), Analysis of hydrographic and stable isotope data to determine water masses, circulation, and mixing in the eastern Great Australian Bight, Journal of Geophysical Research, 144, C l0016, 14pp.

Ridgway, K. R. and S. A. Condie (2004), The 5500-km-long boundary flow off western and southern Australia, Journal of Geophysical Research-Oceans, 109(C4), article C04017.

Rintoul, S. R. and J. L. Bullister (1999), A late winter hydrographic section from Tasmania to Antarctica, Deep-Sea Research I, 46, 1417-1454.

Rintoul, S. R. and S. Sokolov (2001), Baroclinic transport variability of the Antarctic Circumpolar Current south of Australia (WOCE repeat section SR3), Journal o f Geophysical Research, 106( C2), 2815-2832.

Rintoul, S. R. and M. H. England (2002), Ekman Transport Dominates Local Air- Sea Fluxes in Driving Variability of Subantarctic Mode Water, Journal of Physical Oceanography, 32, 1308-1321.

Rochford, D. J. (1960), The intermediate depth waters of the Tasman and Coral Seas, I. the 27.2 sigma-t surface, Australian Journal o f Marine and Freshwater Research, 11, 127-147.

Rochford, D. J. (1964), Hydrology of the Indian Ocean. III. Water masses of the upper 500 m of the south-east Indian Ocean, Australian Journal o f Marine and Freshwater Research 15, 25-55.

Rochford, D. J. (1968), The continuity of water masses along the western boundary of the Tasman and Coral Seas, Australian Journal o f Marine and Freshwater Research, 79,77-90.

60 Rochford, D. J. (1986), Seasonal-changes in the distribution of Leeuwin Current water off southern Australia, Australian Journal o f Marine and Freshwater Research, 37( 1), 1-10.

SAIMOS (2012), Cruise data, Southern Australian Integrated Marine Observing System, SARDI Aquatic Sciences, West Beach, South Australia, 5022, Australia.

Sarmiento, J. L., N. Gruber, M. Brzezinksi and J. Dunne (2004), High latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 426, 56-60.

Schodlok, M. R and Tomczak, M. (1997), The circulation south of Australia derived from an inverse model, Geophysical Research Letters, 24(22), 2781-2784.

Sokolov, S. and S. R. Rintoul (2000), Circulation and water masses along WOCE section PI 1: Papua New Guinea to Tasmania, Journal of Marine Research, 58, 223-268.

Speich, S., B. Blanke, P. de Vries, S. Drijfhout, K. Döös, A. Ganachaud and R. Marsh (2002), Tasman leakage: a new route in the global ocean conveyor belt, Geophysical Research Letters, 29( 10), 1416, doi: 10.1029/2001GL014586.

Stommel, H. (1979), Determination of water mass properties of water pumped down from the Ekman layer to the geostrophic flow below. Proc. Nat. Acad. Sei. USA, 76, 3051-3055.

Stramma, L. (1992), The South Indian Ocean Current, Journal of Physical Oceanography, 22, 421-430.

Sverdrup, H. W., M. W. Johnson and R. H. Fleming (1942), The Oceans, Their Physics, Chemistry and General Biology, Prentice-Hall, Englewood Cliffs, N. J., 1087pp.

Thompson, R. O. R. Y. and R. J. Edwards (1981), Mixing and Water-Mass Formation in the Australian Subantarctic, Journal o f Physical Oceanography,

61 11, 1399-1406.

Thompson, P. A., K. Wild-Allen, M. Lourey, C. Rousseaux, A. M. Waite, M. Feng and L. E. Beckley (2011), Nutrients in an oligotrophic boundary current: Evidence of a new role for the Leeuwin Current, Progress in Oceanography, 97,345-359.

Tomczak, M. (1999), Some historical, theoretical and applied aspects of quantitative water mass analysis, Journal of Marine Research, 57, 275-303.

Tomczak M. and J. S. Godfrey (2003), Regional oceanography: An Introduction, 2nd Edition, Daya, Delhi, 390pp.

Tomczak M. and D. Hao (1989), Water masses in the thermocline of the Coral Sea, Deep-Sea Research, 36(10), 1503-1514.

Tomczak, M. and D. G. B. Large (1989), Optimum Multiparameter Analysis of Mixing in the Thermocline of the Eastern Indian Ocean, Journal of Geophysical Research, 94{C 11), 16,141 -16,149.

Tomczak, M., L. Pender and S. Liefrink (2004), Variability of the Subtropical Front in the Indian Ocean south of Australia, Ocean Dynamics, 54, 506-519.

Vaux, D. and Olsen, A. M. (1961); Use of drift bottles in fisheries research, Australian Fisheries Newsletter, 20, 17-20.

Ward, T. M., S. J. Sorokin, D. R. Currie, P. J. Rogers, L. J. McLeay (2006a), Epifaunal assemblages of the eastern Great Australian Bight: Effectiveness of a benthic protection zone in representing regional biodiversity, Continental Shelf Research 26, 25^40.

Ward, T. M., L. J. McLeay, W. F. Dimmlich, P. J. Rogers, S. A. M. McClatchie, R. Matthews, J. Kämpf and P. D. van Ruth (2006b), Pelagic ecology of a northern boundary current system: effects of upwelling on the production and distribution of sardine (Sardinops sagax), anchovy (Engraulis australis) and southern bluefin tuna (Thunnus maccoyii) in the Great Australian Bight,

62 Fisheries Oceanography, 15(3), 191-207.

Warren, B. A., H. Stommel and J. C. Swallow (1966), Water masses and patterns of flow in the Somali Basin during the southwest monsoon of 1964, Deep-Sea Research, 13, 825-860.

Williams, R. G., M. A. Spall and J. C. Marshall (1995), Does Stommel’s mixed layer “demon” work? Journal of Physical Oceanography, 25, 3089-3102.

Woo, M. and Pattiaratchi, C. (2008), Hydrography and water masses off the western Australian coast, Deep-Sea Research /, 55, 1090-1104.

Wood, J. D. and E. Terray (2005), Ocean current, temperature, and wave observations at the Amrit-1 location offshore of southern Victoria, Australia. Final Report April-October 2004, Ocean Data Technologies Inc. technical report prepared for Santos, Ltd. and Unocal Corporation, 84 pp.

63 Chapter 3. Water masses and their seasonal variation on the Lincoln Shelf, South Australia

Abstract

Five water masses are defined for the Lincoln Shelf, South Australia using hydrographic and stable isotope data collected between 2008 and 2011. Three water masses with distinct source regions or modes of formation are present on the shelf and slope year round. Slope Water with low temperature, salinity and isotope values is present perennially on the slope at depths greater than 180 m, and is episodically upwelled onto the shelf during summer. During strong upwelling events this water can be upwelled from 300 to 100 m water depth on the shelf south of Kangaroo Island. This water then flows west towards Eyre Peninsula and into the mouth of Spencer Gulf. Subtropical Surface Water is a mixed water mass on the shelf and is transported year-round by the eastward flowing South Australian Current. Local heating and evaporation of this water mass on the shelf and within Spencer Gulf during summer forms Evaporated Water, a water mass with high temperature, salinity and isotope values. The use of stable isotopes in water mass analysis has permitted the identification of two new water masses that form on the shelf during summer. Mixed Slope Water is formed when Slope Water mixes with Subtropical Surface Water during upwelling events. Cooled Evaporated Water is generated when surface Evaporated Water mixes vertically with cool, fresh bottom waters. Identification of these mixed water masses has implications for mixing of nutrient- rich mesotrophic upwelled waters with oligotrophic surface waters, a situation that supports greater levels of primary productivity on the shelf.

64 3.1. Introduction

The Lincoln Shelf, South Australia, between Kangaroo Island and the Eyre Peninsula (Figure 3.1), is an area of complex oceanography where several different water masses and currents interact. During winter, the shelf is influenced by the South Australian Current flowing east (Ridgway and Condie, 2004), and highly saline outflow from Spencer G ulf (Lennon et al., 1987) and the Great Australian Bight (Petrusevics et al., 2009). During summer the shelf is influenced by upwelling o f nutrient-rich slope waters (Ward et al., 2006). The Lincoln Shelf is also part of the world’s largest latitude parallel carbonate shelf, which extends from Cape Leeuwin to Tasmania (James et al., 1994) and is globally significant both for ecological reasons and for its role in carbon cycling. Biological production is linked to oceanographic conditions in the region (James et al., 2001) and so greater understanding o f local conditions, especially chemical oceanography and water mass distribution, is important for understanding the productivity of one of the world’s largest carbonate provinces.

South Australia

Eyre Peninsula

s Spencer r% Gulf Gulf St. k SGOi r Vincent Lincoln Shelf

Upwelling Kangaroo Is.

de Couedic Canyon Bonney Coast

Longitude (°E)

Figure 3.1. Lincoln Shelf study area with local currents. Winter South Australian Current (SAC) - warm and relatively saline; Spencer G ulf Outflow (SGO) - cool, saline and dense; and summer upwelling - cool and fresh.

65 3.1.1. Regional circulation and water masses

Only a handful of oceanographic studies have focused on the region and adjacent Great Australian Bight (GAB), and most hydrographic studies have analysed data from specific events or voyages. Modelling studies have identified the seasonal differences in shelf and slope circulation between summer and winter (Middleton and Platov, 2003; Cirano and Middleton, 2004), but observational hydrographic studies using water chemistry are needed to augment and refine these physical modelling studies. This study presents longer term hydrographic and stable isotope data collected over three years from twenty-four voyages, and aims to identify water masses and oceanographic seasonality of the Lincoln Shelf.

The Lincoln Shelf has two distinct seasons, a winter-spring season (June to October) and a summer-autumn season (December to April). During winter-spring, a well- mixed water column is present on the shelf and westerly winds favour downwelling (McLeay et al., 2003). During winter, the Leeuwin Current (Cresswell and Golding, 1980) is the dominant surface current in the region. It is a warm, oligotrophic eastern boundary current that flows poleward from tropical northwest Australia and then east across the southern Australian shelf. It is strongest in this season and reaches as far as the eastern GAB, where it mixes with locally formed shelf water and forms the South Australian Current that flows towards Tasmania (Ridgway and Condie, 2004; Cirano and Middleton, 2004). On the Lincoln Shelf, the South Australian Current flows to the east with speeds up to 40 cm/s over the shelf break (Cirano and Middleton, 2004; Ridgway and Condie, 2004). There is limited information about the South Australian Current and it has yet to be defined hydrographically.

During summer-autumn, waters are stratified on the shelf with strong surface heating and evaporation resulting in a seasonal thermocline at ~50 meters water depth (mwd). Water temperatures can be as high as 22°C within the surface layer, and less than 15°C below the thermocline (Hahn, 1986). Upwelling events can occur 2-4 times a season when winds are upwelling favourable, bringing cold, fresh and nutrient-rich water to the shelf (Middleton and Bye, 2007). Modelling of circulation in the GAB and the Lincoln Shelf during summer shows that wind-forced coastal currents are to the northwest, with an eastward current over the shelf break (Middleton and Platov, 2003). The eastward current occasionally reverses direction

66 during strong summertime upwelling events (John Middleton pers. com., July 5, 2013) The Leeuwin Current is weaker during summer, and is observed only as far east as Esperance around 121°E (Cresswell and Peterson, 1993). Therefore, the source of surface waters in the region is unclear during summer, as is the influence of the South Australian Current.

Below the eastward flowing surface current is the westward flowing Flinders Current, a northern boundary current first identified by Bye (1972; 1983). Modelling studies suggest that the current is strongest in summer and weakest in winter, with maximum speeds of 15 cm/s at -600 m water depth (Middleton and Cirano, 2002; Middleton and Platov, 2003). It transports Subantarctic Mode Water and Antarctic Intermediate Water west along the continental slope, but current meter and hydrographic studies are limited (Hufford et al., 1997; Barker, 2004). New research on the water masses along the continental slope between Tasmania and Cape Leeuwin (Chapter 2 of this thesis) confirms that the Flinders Current transports Subantarctic Mode Water and Antarctic Intermediate Water west along the slope between 400 and 1000 mwd, and that the current also transports South Australian Basin Central Water at depths between 200 and 400 mwd. Therefore, deep upwelling events off Kangaroo Island and the Bonney Coast can be sourced from the Flinders Current.

There is little published hydrographic data for the southern Australian shelf and previous water mass studies are limited. James and Bone (2011) identify surface water along the southern margin as Subtropical Surface Water (STSW) and its definition is quite general, with temperature and salinity ranging from 10-22°C and 35.1-35.9, and an intermediate dissolved oxygen content of 220-245 pmol/L. This water mass is identified as South Indian Central Water on the Western Australian shelf, sourced from the Indian Ocean Subtropical Gyre (Condie and Dunn, 2006; Woo and Pattiaratchi, 2008). On the Lincoln Shelf, Hahn (1986) identified three water masses interacting at the mouth of Spencer Gulf, which were named based on their origin: ocean water, bight water and gulf water. Petrusevics (1993) and Petrusevics et al. (2011) later expanded on these three water masses. Ocean water was cool (<15°C) and fresh (<35.5) and was found below the seasonal thermocline in summer and off shelf during winter. Bight water was warm (18-19°C) and

67 relatively saline (36.0-36.5), forming a low-density tongue of water at the mouth of Spencer Gulf. Bight water was present above the seasonal thermocline in summer and through the whole water column in winter. Gulf water was 2-3°C warmer than bight water (21-22°C), with high salinity (>36.5), and was found within Spencer Gulf.

This study aims to identify water masses for the whole of the Lincoln shelf, identify an upwelled water mass, and better define mixing and circulation between water masses by using stable isotopes of seawater as conservative tracers. The use of stable isotopes of oxygen and hydrogen in conjunction with hydrographic data permits the identification of source regions of water masses and the formation of mixed water masses, which hydrographic data alone cannot provide.

3.1.2. Isotopes in subtropical regions

Stable isotopes of seawater act as more conservative tracers than temperature and salinity, and when water masses mix, isotopic signatures can be retained for a longer period of time. ö'*0 and ö:H values refer to the ratios of 180/l60 and 2H/'H in seawater, which are primarily controlled by evaporation, precipitation and freshwater input (Epstein and Mayeda, 1953; Lloyd, 1966; Gat et al., 1996; Frew et al., 2000). Evaporation preferentially removes the lighter isotope from seawater, increasing seawater isotope ratios, and precipitation and freshwater input add the lighter isotope to seawater, decreasing isotope ratios. The isotopic composition of freshwater and precipitation is also dependent on latitude. Precipitation at higher latitudes has lower isotopic ratios than those at the equator, based on the preferential rain out of the heavy isotope as air masses move poleward (Dansgaard, 1964). Once a water mass is removed from the sea surface its isotopic composition is changed only by mixing. As a result, a water mass will have a distinct isotopic signature based on surface conditions in its source region, as well as transport pathways and mixing patterns as it moves away from the source region. However most isotope studies have focused on oxygen isotope-salinity relationships (Archambeau et al., 1998; Meredith et al., 1999; Khatiwala et al., 1999; Frew et al., 2000; Harwood et al., 2008; McConnell et al., 2009); oxygen-hydrogen isotope relationships and hydrogen isotope-salinity relationships are still poorly understood with respect to latitudinal variation, differences with source regions, and mixing in the ocean.

68 Stable isotope studies in shelf regions have usually been undertaken in areas with large freshwater influx, where large salinity variations provide clear isotope-salinity relationships (e.g. Fairbanks, 1982; Khatiwala et ah, 1999; Povinec et ah, 2008). In comparison, only a handful of studies have focused on subtropical shelf regions, where waters have been isotopically enriched by evaporation and where mixing mainly involves seawater alone (Craig, 1966; Gat et ah, 1996; McConnell et ah, 2009). There is almost no stable isotope data of seawater around Australia, except for two studies in the region around the Lincoln Shelf. Corlis et ah (2003) found non-linear relationships between salinity, öl80 and 62H in Spencer Gulf, Australia. They attributed these relationships to a combination of evaporation, internal mixing and ocean exchange. Richardson et ah (2009) analysed isotope ratios for the eastern GAB and found large variations in ratios due to mixing of shelf waters, upwelled waters and highly evaporated waters formed locally at the head of the Bight. Based on these few studies, it is clear that complex relationships govern isotopes in seawater in evaporative regions with limited freshwater input. In such areas, evaporation is a dominant process but mixing and upwelling also result in significant variability. The conservative nature of isotopes makes them useful proxies that can be used with temperature and salinity to better understand water mass interactions and mixing in shelf regions with complex oceanography, such as on the Lincoln Shelf.

3.2. Methods

This paper analyses a large dataset of temperature and salinity and a smaller dataset of 400 stable isotope analyses. Hydrographic data and water samples for stable isotope analysis were collected aboard the RV Ngerin as part of the Southern Australian Integrated Marine Observing System (SAIMOS; SAIMOS, 2012) (see Appendix C). Research voyages were undertaken on average eight times a year between early 2008 and early 2011 on the Lincoln Shelf between Eyre Peninsula, Kangaroo Island and the mouth of Spencer Gulf (Figure 3.2). Not all CTD stations were sampled during each voyage because of weather conditions and the time of year, but CTD and water samples were collected routinely at biological stations (Figure 3.2b). This dataset is augmented with historic temperature and salinity data collected prior to 2000 accessed through the Commonwealth Scientific and

69 Industrial Research Organisation (CSIRO, 2012). These data were used to create the CSIRO Atlas of Regional Seas (CARS; Ridgway et al., 2002).

All references to density p(S,T,p) throughout this chapter refer to potential density with a reference pressure of p = 0 m (p(S,T,0)). In-situ temperature rather than potential temperature was used to calculate potential density, therefore compression effects but not adiabatic effects were accounted for. However, for the data presented here, the errors relating to this are less than 0.3% (see Appendix B). For convenience, potential density is reported as sigma-t (at) throughout this chapter, and is defined as:

ot = p(S,T,0) - 1000 kg/m3

The equation of state used in this chapter is UNESCO EOS-80 using practical salinity units. Pressure data is here reported as depth in meters, as 1 m is approximately equal to 1 decibar (within 1% error for the depth range 0-1200 m).

Water samples for stable isotope analysis were collected using Niskin bottles at three depths - 15 mwd, the deep chlorophyll maximum ~50 mwd, and ~10 m above the sea floor. Samples for isotope analysis were collected between August 2008 and February 2011 predominantly from water depths less than 110 m. Water samples were measured for oxygen and hydrogen isotopic compositions at the Queen’s Facility for Isotope Research in Kingston, Canada. Oxygen isotopic compositions were measured using a modified CO2-H2O equilibration technique of Epstein and Mayeda (1953) on a Finnigan GasBench II inline with a Finnigan DELTAplusXP stable isotope mass spectrometer. Hydrogen isotopic compositions were measured on a Finnigan MAT 252 mass spectrometer connected to an automated Finnigan H- Device. Measurements of 2H/'H and 180/l(10 ratios are presented in standard “8” notation with respect to Vienna-Standard Mean Ocean Water (V-SMOW). The 8 value is defined as:

8 (%o) — (R/R v -s mow — 1) x 10’ where R is the isotope ratio 2H/'H or l80 /160. Repeated analysis of lab seawater standards showed reproducibility (lo standard deviation) of ± 0.4 per mil and ±0.13 per mil for 82H and 8lsO measurements, respectively.

70 (a)

Eyre Peninsula Spencer Gulf St. ■ Gulf Vincent

100m ' 120m ■ 600m,

Kangaroo Is.

x ' ’

• Historic CTD locations x SAIMOS CTD locations

Longitude (°E) (b)

GAB •

Lincoln Shelf

* *

O Biological stations * Isotope stations

Longitude (°E)

Figure 3.2. Lincoln Shelf study area between Eyre Peninsula and Kangaroo Island, showing station locations, a) CTD stations of SAIMOS sampled between February 2008 and February 2011 (blue crosses) and historic data from CARS data, sampled prior to 2000 (red points), b) Routine biological stations sampling isotopes (pink circles) and isotope stations sampled during select months (black stars).

71 3.3. Summer Water Masses

3.3.1. Introduction

Two seasonal states are present on the Lincoln Shelf, a ‘summer’ state from December to April and a ‘winter’ state from June to October. May and November appear to be transition months on the shelf. During May, shelf data matches summer data, however gulf data shows cooling and a transition between summer and winter gulf water properties. During November, shelf data retains winter properties if it precedes a weak upwelling season and has summer properties if it precedes a strong upwelling season.

Waters on the Lincoln Shelf during summer months are stratified with a surface mixed layer of 0-40 mwd and a seasonal thermocline between 40-60 mwd. During upwelling events, this surface mixed layer can be as shallow as 20 mwd around Kangaroo Island. A seasonal halocline coincides with the thermocline on the inner shelf, but it is less pronounced on the outer shelf where surface waters are fresher. Salinity variations between surface and bottom waters can be as high as 0.8 psu on the inner shelf, as was found on the shelf adjacent to the mouth of Spencer Gulf during January 2010.

To determine water masses, we first present some basic concepts of water mass analysis and consider the temperature-salinity (T-S) diagram (Figure 3.3) for all of the available CARS and SAIMOS summer data. Three end members are identified and form a mixing ‘triangle’, a cool-fresh end member (1), a warm-fresh end member (2) and a warm-saline end member (3). Between end members 1 and 2, the T-S values form a linear trend indicating mixing between these two end members. The line also cuts across density contours, indicating that diapycnal mixing has occurred. Approximate straight lines between end members 2 and 3, as well as 3 and 1, also indicate mixing between these end members. The scatter of the data is a result of mixing between all three end members.

The relationship between 5lxO and 8:H (Figure 3.4a) shows only two end members rather than three. During mixing, heat diffuses faster than dissolved ions, and during evaporation salinity is primarily affected. As temperature and salinity react differently, they can change independently of each other and form multiple end

72 members. In comparison, isotopes in seawater are generally modified in the same way by mixing and evaporation, resulting in a single mixing line between low isotope values and high isotope values. Using isotopes as the most conservative tracers, water masses can be defined along the single mixing line of 5I80 vs 8: H. These water masses can then be plotted in T-S space (Figure 3.4b), providing additional information to the T-S results in Figure 3.3 and 3.4a, or validating existing information such as source region, spatial variability and paths of mixing.

(2) warm, fresh f .. >18°C, <35.6

(1) cool, fresh -<13°C, <35.3 X Shelf (50-250 mwd) X Gulf (<50 mwd)

Figure 3.3. Temperature-salinity plot for the summer season (months December-May), for the shelf in blue (50-250 m), and gulfs in red (less than 50 m) (CARS and SAIMOS data). Grey lines denote density. Plot shows three summer end members: cool, fresh end member 1, warm, fresh end member 2 and warm, saline end member 3. The data plot as a triangle between these end members.

Five new water masses are defined on the shelf during summer, firstly by their isotopic compositions and location along the isotope mixing line and secondly by salinity, temperature and spatial distribution (Figure 3.4). Three water masses are found to have distinct source regions or formation modes, Slope Water (SW), Subtropical Surface Water (STSW) and Evaporated Water (EW), and two are a result o f mixing or modification of the original three, resulting in Mixed Slope Water (MSW) and Cooled Evaporated Water (CEW). Water mass properties are summarised in Table 3.1. A ll five water masses can be present on the shelf during summer and a schematic o f their spatial distribution is given in Figure 3.5.

73 (a) Summer isotope water masses - ö180 vs 5 2H

* SW 0 MSW x STSW □ EW a CEW

ö180 (per mil)

(t>) Temperature vs Salinity by water mass

(3 - EW) (2 - MSW)

* SW 0 MSW x STSW □ EW A CEW

Salinity

Figure 3.4. Summer 6“H vs 5 lxO (a) and temperature vs salinity (b) o f isotope samples by water mass (defined by isotopes, salinity and temperature). 2a) shows two isotope end members, 1) low isotope Slope Water (SW) and 2) high isotope Evaporated Water (EW). Mixed Slope Water (MSW), Subtropical Surface Water (STSW) and Cooled Evaporated Water (CEW) are on the mixing line between these two end members. 2b) shows the three T-S end members seen in Figure 3.3, and associates them with water masses: 1) cool, fresh SW 2) warm, fresh MSW and 3) warm, saline EW. Black arrows show mixing directions.

74 Table 3.1: Summer water mass properties, defined by T, S and isotopes.

W ater mass Temperature (°C) Salinity 6 lxO (%o) 5 2H (% o )

Slope Water (SW) 10.4-14.7 34.85-35.55 -0.05-0.32 1.2-2.6

Mixed Slope Water (MSW ) 13.4-19.4 35.3-35.9 0.05-0.37 2.2-3.2

Subtropical Surface Water (STSW) 13.3-20.1 35.35-36.2 0.08-0.55 2.6-4.2

Evaporated Water (EW ) 17.5-21.3 35.7-37.3 0.29-0.86 3.5-7.4

Cooled Evaporated Water (CEW) 14.1-16.8 35.4-36.2 0.35-0.54 3.6-6

South Eyre Peninsula Australia

Kangaroo Island

Longitude Eyre Peninsula Kangaroo Island

Evaporated Water patches

vertical mixing Subtropical Surface Water double diffusion

Cooled Evaporated Water westward flow, mixing

- 100m Mixed Slope Water Slope Water Kangaroo Island Upwelling I ...... i 135.5°E Longitudinal Section 136.5°E

Figure 3.5. Schematic o f summer shelf water masses during periods o f upwelling for section A -A l (see top inset for location). Slope Water mixes with Subtropical Surface Water to form Mixed Slope Water as it flows west and mixes vertically on the shelf. Patches o f Evaporated Water form at the surface from surface heating and evaporation, and vertical mixing and possible double diffusion processes mix it with bottom waters to form Cooled Evaporated Water.

75 3.3.2. Slope Water

Definition and origins

The Slope Water (SW) end member has low isotope values (S180 <0.32%o, 52H <2.6%o), low temperature (<14.7°C) and low salinity (<35.55) (Figure 3.4; Table 3.1). This water mass is found year-round at -180 mwd on the outer shelf and upper slope and is classified as upwelled water when found on the shelf during summer. Upwelling from depths >180 mwd brings SW onto the shelf.

Temperature, salinity, 5180 and 52H values of SW on the shelf were as low as 10.4°C, 34.85, -0.07%o and 1.2%o, respectively, recorded at a depth of 105 mwd southwest of Kangaroo Island during March 2010. These properties, as well as plots of the data with depth, show that SW can be upwelled from depths of -300 m during strong upwelling events, bringing South Australian Basin Central Water (SABCW) onto the shelf to depths o f-100 m (Chapter 2, Chapter 4 of this thesis). SABCW has temperature and salinity characteristics between 10 and 12°C and 34.8 and 35.1, respectively, and 5180 and 82H values less than 0.1 %o and less than 2%o, respectively. It is formed at the Subtropical Front within the South Australian Basin south of Australia and is transported north and then west along the continental slope by the Flinders Current at depths of -200 to 400 m (Chapter 2 of this thesis).

Spatial distribution

SW is only present on the shelf during upwelling events. During periods of weak upwelling, such as in January 2009, March 2009 and January 2010, SW was confined to the shelf edge at near bottom depths south of Kangaroo Island and proximal to large submarine canyons. However, during strong upwelling events, such as February and March 2010, SW is observed across the whole shelf at depth (Figure 3.6), including the mouth of Spencer Gulf and as shallow as 50 mwd south of Kangaroo Island and Eyre Peninsula. Based on the spatial distribution of SW, it appears to be upwelled through submarine canyons to the south of Kangaroo Island, or is upwelled further to the east and then transported into the region (Chapter 4 of this thesis). It spreads west along the shelf bottom towards Eyre Peninsula and north into the mouth of Spencer Gulf, and mixes vertically through the water column.

76 Figure 3.6. Bottom plot o f a) temperature and b) salinity for February 2010, a month of strong summer upwelling. Water masses and mixing direction on a) - Cold and fresh Slope Water (SW) is present close to the shelf edge and west of Kangaroo Island, with Mixed Slope Water (MSW) present to the west, and warm, saline Evaporated Water (EW) present within Spencer Gulf.

3.3.3. Subtropical Surface Water

Definition and origins

Subtropical Surface Water (STSW) has intermediate isotope values (5I80 0.08- 0.55%o, 5:H 2.6-4.2%o), temperature (13-20°C) and salinity (35.35-36.2) (Figure 3.4; Table 3.1). It is referenced in the literature as shelf waters along the southern Australian continental margin (James and Bone, 2011) and has also been called South Indian Central Water from the western Australian coast to the GAB (Condie and Dunn, 2006; Woo and Pattiaratchi, 2008). This water mass is carried east by the South Australian Current onto the Lincoln Shelf. It can also form locally from mixing because its intermediate properties are on the mixing line between SW and EW properties.

Spatial distribution

STSW was found to be the most abundant shelf water and is present year round on the Lincoln Shelf. It is widespread on the shelf from surface to sea floor depths except during upwelling periods when SW and MSW are present at depth, and during intense summer heating and evaporation when it is modified to form EW at the surface (Figure 3.5).

77 3.3.4. Evaporated Water

Definition and origins

Evaporated Water (EW) forms the end member with high isotope values (5I80

>0.29%o, 82H >3.5%o), high temperature (17.5-21.3°C) and high salinity (35.7-37.3) (Table 3.1; Figure 3.4). This water mass is formed locally from surface heating and evaporation on the shallow inner shelf and within Spencer Gulf during summer (Figures 3.5, 3.6).

Spatial distribution

EW is present both at the surface and at depth (40 mwd) within Spencer Gulf early in summer, but as summer progresses it is also found at the surface on the inner and mid shelf. Its presence within Spencer Gulf is consistent throughout summer and between years, but its presence on the shelf is temporary as its high isotope signal is lost when mixing occurs with shelf waters of lower isotopic value, such as STS W. It is most common on the shelf late in the summer during non-upwelling years, such as during February 2009, when shelf waters are not mixing with SW and MSW of lower isotopic value.

3.3.5. Mixed Slope Water

Definition and origins

Mixed Slope Water (MSW) is a warm, fresh end member (Figure 3.3) and is a mixed water mass with low to mid isotope values (5I80 0.05-0.37%o, 52H 2.2-3.2%o), as well as intermediate temperature (13-19°C) and salinity (35.3-35.9) (Figure 3.4; Table 3.1). It is formed when SW is brought onto the shelf during upwelling events and early mixing occurs between SW and STSW. Initial mixing results in an increase in temperature and formation of warm, low salinity water (e.g. Figure 3.4b).

Spatial distribution

MSW is spatially distributed between SW and STSW during upwelling events and can be found at all depths. It is only on the shelf when SW is present, except for two samples close to the shelf edge at depth southwest of Kangaroo Island during December 2009. It is predominantly found to the north, west and vertically above

78 SW, suggesting that SW is mixing with STSW in these directions to form MSW (Figure 3.5).

3.3.6. Cooled Evaporated Water

Definition and origins

Cooled Evaporated Water (CEW) has high isotope values (8'*0 0.35-0.54%o, 52H 3.6-6%o) but intermediate temperature (14-17°C) and salinity (35.4-36.2) (Figure 3.4; Table 3.1). CEW has an identical isotopic range to EW but is found at depth rather than at the surface, and has lower temperature and salinity than EW. This water mass is interpreted to be EW that has formed at the surface and has then been transferred to bottom depths by vertical mixing processes. It has a cooler and fresher T-S signature than EW from mixing with bottom waters, but mixing has not progressed enough to change the isotopic signature. CEW is a summer water mass because it is formed from vertical mixing with cool and fresh bottom waters, which are only present in summer. It is only identified during summer months when there is weak or no upwelling. It may still be present during strong upwelling months, however, but its signal is indistinguishable, as upwelled water of low isotopic value mixes and erodes the high isotopic signal of CEW. During summer months with weak or no upwelling, vertical mixing may extend to greater depths and mix surface waters from the surface to bottom. A second possible mechanism for vertical mixing is double diffusion by salt fingering, which has been identified in the region (Doubell et al., 2014). Salt fingering occurs when hot, salty water overlies cool, fresh water. If a finger of hot, salty water penetrates into the colder, fresher water below, it will become denser than its environment and continue to sink, as the heat diffuses more quickly than salt (Turner, 1979). This process will continue as the water loses heat and continues to sink, mixing vertically through the water column. Such a process would be stronger during larger upwelling events, but because CEW is indistinguishable during these times, it is not possible to identify which vertical mixing process is dominant.

Spatial distribution

CEW was most common during February 2009 when there was widespread EW at the surface across the shelf, and CEW was found at sea floor depths below this

79 surface water. It was present to the west and northwest of Kangaroo Island and close to the shelf edge to the east of de Couedic Canyon, south of Kangaroo Island. This bottom water was also present in isolated patches around the western tip of Kangaroo Island in January and November 2009 and January and December 2010. The predominance of this water in February 2009 is attributed to conditions during that month. Summer 2009 had no significant upwelling, which in other years would have brought water with low isotope values onto the shelf and obscured the high isotope value of CEW. Furthermore February was the peak month of heating and evaporation during that summer, enhancing formation of EW and the double diffusion process.

3.3.7. Implications for mixing

During summer upwelling events, all five water masses are present on the shelf. The isotope mixing line shows that SW can be modified to MSW, then to STSW and evaporated further to EW (Figure 3.4). Therefore through mixing, SW on the shelf increases in temperature first to become MSW, and then increases in salinity and heavy stable isotopes to become STSW and EW. During upwelling, diapycnal mixing occurs across density lines as water properties transition from cool-fresh to warm-fresh end members. The downward mixing of water on the other hand, maintains a similar density as it transitions from warm-saline to cool-fresh properties. CEW is predominantly found between warm-saline and cool-fresh end members, and in comparison to MSW, has higher salinity and higher isotope values, an indication of time spent evaporating at the surface.

The spatial distribution of MSW can provide information on the path of upwelled SW on the shelf. During months with minor upwelling, SW is found only south and southwest of Kangaroo Island, but MSW can be present at depth to the west as far as Eyre Peninsula. This relationship suggests that SW is modified to MSW as it flows west towards the GAB. This interpretation is in agreement with previous modelling by Middleton and Platov (2003) who identified northwestward coastal currents during summer. McClatchie et al. (2006) also suggested that water is upwelled south of Kangaroo Island and is transported into the eastern GAB by coastal currents. The presence of MSW at the mouth of Spencer Gulf during major upwelling events also indicates that upwelled water flows north into shallow inner shelf and gulf waters.

80 3.4. Winter Water Masses

3.4.1. Introduction

During winter, the shelf waters are subject to downwelling that arises from the predominantly eastward winds and the formation of relatively dense water in the shallow coastal regions and gulfs. This dense water is formed through winter cooling as well as evaporation, which exceeds precipitation all year round. Typically the downwelling mixes water from the surface to depths of 150 m. Off Kangaroo Island, the outflow of dense water from the gulfs can reach depths of 250 mwd (Middleton and Bye, 2007). The predominantly eastward winds also act to intensify the eastward flowing SAC.

The majority of winter data analysed in this study (June to October) were collected by SAIMOS on five voyages between August 2008 and July 2010. November is defined as a transition month, as it retained winter characteristics during 2008 and 2010, but had summer characteristics in 2009, beginning the strong 2009-2010 upwelling season. Consistent with the above description of the physics, these data indicate a well-mixed water column to depths of -150 mwd and small temperature and salinity variations with depth. A temperature gradient was present across the shelf, with warmer waters close to the shelf edge and cooler waters on the inner shelf. This was due to both atmospheric cooling in shallow waters and the core of the warm SAC over the shelf break.

Temperature and salinity data collected on SAIMOS voyages (Figure 3.7) show three end members, a cool-fresh end member (1), a warm-moderate salinity end member (2) and a cool-saline end member (3). End member 1 is present in off shelf slope profiles, end member 2 includes the majority of winter shelf data, and end member 3 is found in gulf profiles. The data also show monthly and yearly variation; e.g. 1) warmer temperatures in June 2009, and 2) cooler and fresher water in October and November 2009. The latter variation is on the mixing line between end members 1 and 2, a precursor to the 2009-2010 strong upwelling season. October and November 2008, in comparison, precede a weak upwelling season and have data within the range of end member 2.

81 20

X Shelf (50-250m) warm June 2009 X Gulf (<50m) X Off-shelf profiles (>180m) November transition

(2) warm winter mixed water O 16 ~16°C, -35.8

«3 15 Q.

f x y ^ ------/( 3 ) cool, saline $ x N. October 2009 gulf water November 2009 14-15°C, >36.2

(1) cool, fresh / Off-shelf profiles ' *

36 Salinity

Figure 3.7. Temperature vs salinity plot for the winter season (months June to October), for the shelf in blue, and gulfs in red (CARS and SAIMOS data). November data shown in yellow is identified as a transition month depending on the year. Profiles to -3 50m from directly o ff the shelf (water depths greater than 180 m) are shown in black. Grey lines denote density. Plot shows three end members; winter shelf data (in blue) also includes warm water during June 2009, and some cooler water in October 2009 preceding a strong upwelling season.

During winter, only three of the five summer water masses identified above are present in the region (Table 3.2). Based on isotope data (Figure 3.8), end member 1 relates to S W and is found on the slope at depths greater than 180 m, end member 2 relates to STSW on the shelf, and end member 3 relates to EW both on the shelf and within the gulf. Due to the presence of these three water masses year round, the range o f isotope values is similar for summer and winter (Figure 3.9), although there are significantly more samples with lower values during summer due to the presence of SW on the shelf. Isotopes do however vary considerably for a given salinity range, and this is highlighted during winter when salinity levels are fairly uniform on the shelf, as shown in Figure 3.9b for 82H versus salinity at -35.8-36.0: isotope values range from 2.7 to 6.2%o.

8 2 Winter isotope water masses - ö180 vs ö2H 8

EW

□ pa t XXX xin 0 □D X X XK X X X X X I“ X X XSK X X X X XK X X x xxx x x XK X XK X ^XK X X

XK X X X

SW * SW x STSW □ EW

- 0.2 0.2 0.4 0.6 0.8 ö180 (per mil)

(b ) Temperature vs Salinity by water mass

(2 - STSW)

(3 - EW)

(1 - SW) * SW x STSW □ EW

Salinity

Figure 3.8: Winter 62H vs ÖlxO (a) and temperature vs salinity (b) o f isotope samples by water mass (defined by isotopes, salinity and temperature). STSW and EW can have a similar T-S signature during winter but are distinct isotopically. Numbers denote T-S end members o f Figure 3.7 and correspond to slope, shelf and gulf regions respectively.

83 Table 3.2: Winter water mass properties, defined by T, S and isotopes.

Water mass Temperature (°C) Salinity Sl80 (%o) 62H (% o )

Slope Water (SW) (n = 3) <13.5 <35.3 0.04-0.14 1.4-2.3

Subtropical Surface Water (STSW) 13.9-18.3 35.35-36.2 0.15-0.48 2.7-4.5

Evaporated Water (EW) 13.7-18.3 35.7-37.1 0.47-0.74 4.0-6.2

STSW

-0.2 0 0.2 0.4 0.6 0.8 1 34.5 35 35.5 36 36.5 37 37.5 ö,80 (per mil) Salinity

Figure 3.9. Summer and winter values for (a) 5~H vs 8lxO and (b) 6:H vs salinity. Summer values in blue x and winter values in black *. The three lowest isotope values during winter in (a) were sampled in water depths greater than 180 m from off shelf locations.

3.4.2. Slope Water

Definition

Only three samples of SW were identified in the SAIMOS data, due to limited sampling close to the shelf edge and off the shelf during winter. SW during winter is within the range of summer SW, with low isotope values (5l80 <0.14%o, 52H <2.3%o), salinity (<35.3) and temperature (<13.5°C) (Figure 3.8; Table 3.2).

84 Spatial distribution

The three samples of winter SW are from depths greater than 180 mwd, therefore SW does not intrude onto the shelf during this season. This is because westerly winds promote downwelling in the region (McLeay et ah, 2003), thus restricting SW to the upper slope.

3.4.3. Subtropical Surface Water

Definition

STSW during winter has identical isotope values (5lxO 0.15-0.48%o, S2H 2.7-4.5%o) and salinity (35.35-36.2) to summer STSW, but has a slightly smaller temperature range (13.9-18.3°C) (Tables 3.1, 3.2). STSW shows no change in source water properties between seasons and this characteristic suggests that the South Australian Current flows year round, transporting STSW from the west. The South Australian Current does not show a change in source region signature with each season, like the Leeuwin Current does south of Western Australia when it changes from tropical in winter to subtropical in summer (Cresswell and Peterson, 1993). This consistency may be due to the predominant influence of GAB water and local modifications more than Leeuwin Current water, which varies in strength over the year and does not flow this far east (Ridgway and Condie, 2004; Cirano and Middleton, 2004). The T-S signature of the South Australian Current is identified as the grouping of winter mixed water samples in Figure 3.7 at ~16°C/35.8.

Spatial distribution

STSW is found across the whole shelf during winter. Several samples from 150 mwd are identified as STSW, signifying a well-mixed water column from the surface to this depth. The T-S signature of the South Australian Current is most pronounced on the mid and outer shelf where the current is probably strongest.

3.4.4. Evaporated Water

Definition

EW during winter is distinguished by high salinity (35.7-37.1) and high isotope values (5180 0.47-0.74%o, 52H 4.0-6.2%o) but has cooler temperatures than during

85 summer (13.7-18.3°C) (Figure 3.8; Table 3.2). Winter EW is most likely a remnant of summer EW and may also be sourced from the GAB, where similar evaporated water would have formed in summer from strong surface heating and evaporation. The South Australian Current would then transport this evaporated water into the region.

Spatial distribution

EW is found on the shelf and within Spencer Gulf during winter. In contrast to its defined surface distribution in summer, in winter EW can be found at all depths amongst STSW on the shelf, and does not show a strong spatial pattern, suggesting that the shelf during winter is not well mixed isotopically. There is however a predominance of deep EW samples west and south of Kangaroo Island; in conjunction with high salinity this water is identified as winter outflow from Spencer Gulf.

Spencer Gulf Outflow

Spencer Gulf Outflow (Godfrey et al., 1986; Petrusevics et al., 2009) is identified as EW on the shelf at depth around the western tip of Kangaroo Island during winter. T-S properties show diapycnal mixing between STSW and EW (Figures 3.7, 3.8). The outflow is distinguished in bottom contour charts (Figure 3.10) with salinity and density as high as 36.5 and 27.8 kg/m’ on the shelf.

Figure 3.10. Bottom contours of a) salinity and b) density during October 2008. The outflow is seen as a low temperature, high salinity, high density plume. STSW - Subtropical Surface Water, EW - Evaporated Water, WI - Wedge Island.

86 Spencer Gulf outflow predominantly exits from southeastern Spencer Gulf, off York Peninsula. A smaller high salinity plume can also exit to the west of Wedge Island in the centre of the gulf mouth, as observed during October 2008 (Figure 3.10). On the slope, highest recorded isotope values were found at 150-160 mwd during June 2009, an indication of outflow reaching the shelf edge. Spencer Gulf outflow was present from June to October in all SAIMOS data. It was also present in November 2008 and 2010, and just south of the gulf mouth in April 2011. However, upwelling seasons restricted the temporal extent of outflow; it was absent in April 2008 and 2010 due to strong upwelling events in late summer and the persistence of summer conditions, and in November 2009 due to the onset of the strong 2009-2010 upwelling season. The month of May was not sampled during SAIMOS voyages, which may be when the outflow is first apparent.

3.4.5. Implications for mixing

The larger dataset of temperature and salinity shows mixing between STSW and EW when Spencer Gulf Outflow is on the shelf, and mixing between STSW and SW in late spring or early summer (Figure 3.7). In 2008, cool, fresh water did not begin to mix with shelf water until December, however the following year mixing was identified as early as October. Despite this, isotope values do not indicate SW on the shelf before December 2009; it may be that any influx of cooler, fresher water was coming from shallower than 180 mwd and not sourcing SW of different isotopic composition. The cooler and fresher water present on the shelf during October and November 2009 is still significant however, as it was a precursor to the strong upwelling season of summer 2009-2010. Appearance of this cooler, fresher water during late spring could suggest a strong upwelling season to come. In comparison, gulf outflow and winter conditions were still observed on the shelf during November 2008, and this month preceded the weak upwelling season of 2008-2009.

3.5. Discussion

3.5.1. Temporal and seasonal variation

In this study, the distribution of water masses determined whether a month was categorised as a winter month or a summer month. The summer months were categorised further as having no upwelling, weak upwelling or strong upwelling

87 based on the presence of SW and MSW. Months of strong or weak upwelling were dominated by SW and MSW at depth and had lower isotope values. SW was present during all four summer seasons sampled, with no SW prior to January and stronger upwelling events later in the summer, from February onwards. February and March 2008 and February and March 2010 were strong upwelling months and show SW at depth across the shelf. March 2010 was the strongest recorded upwelling month, with the lowest isotope value on the shelf and the second lowest value overall at 110 mwd southwest of Kangaroo Island. March 2009, January 2010, April 2010 and February 2011 were weak upwelling months and had MSW across the shelf, with SW restricted to south of Kangaroo Island or close to the shelf edge. Summer months with no upwelling show STSW with possible CEW at depth rather than SW. No significant SW was found on the shelf during November and December 2008, January, February, November and December 2009, and November and December 2010.

Summer months with no upwelling were isotopically similar to winter months. This suggests that without any slope water influence, surface waters have the same source year round, implying the influence of an eastward flowing South Australian Current during both winter and summer. A strong eastward flowing current during winter is well documented, and Ridgway and Condie (2004) defined the South Australian Current as eastward winter flow between the GAB and Bass Strait. Less attention had been given to such a current during summer, however. Until now, questions still remained about the source of surface water present on the Lincoln Shelf during summer when the Leeuwin Current and South Australian Current are weaker. Middleton and Platov (2003) modelled an eastward flow over the shelf break during summer, but as yet no observational data had been published. New unpublished current meter data from a SAIMOS mooring south of Eyre Peninsula recorded between October 2009 and November 2010 shows that on average, currents are flowing to the east, with short periods of westward flow during summer (SAIMOS, 2012). The summer South Australian Current could be an extension of the southeasterly flowing Great Australian Bight Plume identified by Herzfeld and Tomczak (1999), Middleton and Platov (2003) and Richardson et al. (2009) in the GAB. This water mass study herein supports the limited documentation of an

88 eastward flowing current during summer as well as winter, and the predominance of STSW year-round in the region.

The source of STSW and the eastward flowing current is also supported by comparison of T-S data between the eastern GAB and Lincoln Shelf using CARS and SAIMOS data (not shown). These two regions are very similar throughout the year. Only during months with strong upwelling around Kangaroo Island is water on the Lincoln Shelf cooler and fresher than the GAB region. However these regions are comparable during summer months when there is little to no upwelling around Kangaroo Island and STSW is predominant on the Lincoln Shelf. Both regions also have warm, saline water; waters in shallow areas of the GAB match well with waters in the shallow mouth of Spencer Gulf. Based on such similar characteristics, it is likely that these two regions are influenced by the same water mass, carried by the South Australian Current, and are affected by similar climatic conditions, such as strong heating and evaporation during summer.

3.5.2. Comparison with previous water mass studies

Water masses defined in this study differ somewhat from those recognised in previous studies of the Lincoln Shelf. Hahn (1986) first identified three water masses at the mouth of Spencer Gulf, which were then confirmed and expanded on by Petrusevics (1993) and Petrusevics et al. (2011). The current study has a larger spatial extent, from Spencer Gulf to the shelf edge south of Kangaroo Island, and incorporates stable isotopic compositions into water mass definitions. The use of isotope data in water mass analysis has allowed water masses to be defined by source region as well as hydrographic properties.

Gulf water as defined by Hahn (1986) with high temperature and salinity is similar to EW from this study, formed by summer heating and evaporation. EW is present both in the gulf and in surface waters on the shelf during some summer months, however, and was found down to 60 mwd on the mid shelf during February 2009 when EW was widespread at the surface. Therefore these highly evaporated waters can form at the surface on the mid shelf and are not restricted to gulf waters. A second water mass defined by Hahn (1986) was ocean water, and was found below the seasonal thermocline during summer with low temperature and salinity. In comparison, cool and fresh SW is defined in this study as water coming from the 89 slope at depths below the oligotrophic surface waters, and its presence below the seasonal thermocline is dependent on the occurrence of upwelling events. The SW mass defined here more specifically identifies upwelling that brings nutrient-rich water with low isotope values to the shelf. During periods with no upwelling, water below the seasonal thermocline is isotopically similar to water above it, and is therefore identified as having the same source region. In this study such water is identified as STSW, and comes from the GAB.

Hahn (1986) also defined bight water coming from the GAB, with lower density compared to gulf water and ocean water. This density minimum is associated with the formation of fronts across the mouth of Spencer Gulf (Petrusevics, 1993; Petrusevics et al., 2011), and is related to warm water from the bight intruding into the mouth of Spencer Gulf. In the current study however, this low-density water does not appear to be widespread on the shelf. A second location of low-density water is found at the shelf edge where lower salinity water is present at the surface, but this is due to lower salinity Southern Ocean water being warmed by surface heating during summer (CARS; Middleton and Bye, 2007). STSW in this study is also defined as having a GAB source like the bight water defined by Hahn (1986), but it is not distinguished by density. Stable isotopic compositions indicate STSW is widespread on the shelf during both summer and winter, and is only displaced by SW during upwelling events and modified to EW by heating and evaporation.

In addition to these three water masses, the analysis of isotope data has allowed the identification of MSW and CEW. MSW provides important spatial information on the path of upwelled nutrient-rich water on the shelf, and CEW suggests that vertical mixing of surface and deep waters occurs during summer, which could bring nutrient-rich water into the photic zone to be utilised by photosynthetic organisms.

3.5.3. Comparison with previous isotope studies

Two previous studies analysed stable isotopes in the region, although this is the first study of waters over the Lincoln Shelf. Corlis et al. (2003) analysed 6lsO and 82H data from summer and winter within Spencer Gulf. The current study overlaps spatially only at the mouth of Spencer Gulf, and 82H data are in general agreement between the two studies. 8lsO on the other hand is lower by ~0.3%o in the current

90 study, possibly due to different climatic conditions or upwelling influence during both periods of study. Isotope-isotope and isotope-salinity relationships are not comparable due to differences between the open shelf and the restricted embayment of Spencer Gulf. A preliminary study of shelf water masses from the eastern GAB during an upwelling event in March 1998 (Richardson et al., 2009) found large variations in both oxygen and hydrogen isotope values that are not found in the current study. The current study in comparison shows good oxygen-hydrogen relationships and mixing lines, and represents either data of higher quality or smaller variations because of the variable extent of mixing that can occur annually on the shelf.

3.6. Conclusions

While there have been studies of marine waters at the mouth of Spencer Gulf, this study is the first to identify water masses for the whole shelf, recognize the origin of upwelled waters, and better define mixing and circulation between water masses. The use of stable isotopes of oxygen and hydrogen together with hydrographic data has made it possible to identify source regions of water masses and the formation of mixed waters, and better define the spatial distribution of water masses on the Lincoln Shelf. Five new water masses are defined on the shelf, including three water masses with specific source regions and formation modes that are present year round, and two water masses that form by variable mixing on the shelf during summer (Figures 3.4, 3.5). The three source water masses are:

• Slope Water (SW), which is a low temperature, salinity and stable isotope end member that is found year round at depths greater than 180 m, and is sporadically upwelled onto the shelf during summer from depths as great as 300 m. This water mass is sourced from South Australian Basin Central Water, which is formed at the Subtropical Front south of Australia and is transported west along the slope by the Flinders Current. It is thought to upwell through submarine canyons to the south of Kangaroo Island or further to the east and transported into the region by a westward flowing shelf current. • Subtropical Surface Water (STSW), which is a mixed shelf water mass with intermediate salinity and stable isotope values that is transported year round

91 into the region by the eastward flowing South Australian Current. It is well mixed on the shelf down to depths of 180 m during winter, and is displaced at depth by the inflow of SW during summer upwelling events. • Evaporated Water (EW), which is a high salinity and stable isotope end member that is locally formed at the surface on the inner shelf and within Spencer Gulf from high evaporation and heating during summer. It outflows from Spencer Gulf onto the shelf at depth during winter.

The use of stable isotopes has made it possible to identify two additional mixed water masses, which are produced on the shelf during summer. The first is Mixed Slope Water (MSW). Despite having moderate temperature and salinity, stable isotope values indicate it is formed by mixing of S W with STS W on the shelf during upwelling events, as it lies on the isotopic mixing line between these two source water masses. Spatial distribution of this water mass in relation to SW suggests that SW is transported west towards Eyre Peninsula and north into the mouth of Spencer Gulf, and vertical mixing allows nutrients of SW to be brought into the photic zone to be utilised by primary producers.

The second mixed water mass is Cooled Evaporated Water (CEW), a modification of EW that is identified from stable isotope values identical to EW, but with lower temperatures and salinities. It is found near the seafloor during summer, predominantly around the western tip of Kangaroo Island, however the identical stable isotope signature to EW confirms that this water mass had recently been evaporated at the surface. Early stages of vertical mixing between surface EW and cooler, fresher bottom waters result in equilibration of temperature and salinity but preservation of the more conservative stable isotope signatures, to form CEW. This recognition of vertical mixing through a stable stratified water column in the region could be important for vertical mixing of nutrients both on the shelf and in Spencer Gulf. A possible mechanism to achieve this phenomenon is double diffusion by salt fingering, whereby hot, salty surface water (EW) can mix with near bottom cool, fresh water. Double diffusion may be a mechanism to bring nutrients from SW and MSW into the photic zone.

The use of stable isotopes as conservative tracers in water mass analysis provides additional information that temperature and salinity alone cannot supply. As isotopes

92 are able to identify the source region of a water mass, surface waters on the Lincoln Shelf, identified as STSW, have the same source region during both summer and winter. The South Australian Current transports STSW from the Great Australian Bight to the Lincoln Shelf, therefore the presence of STSW year round in the region confirms that the current is flowing in summer as well as winter, which has not been previously identified. Despite previously recognised high seasonality in the region, STSW and EW were found on the Lincoln Shelf during summer and winter, and while SW was not present on the shelf during winter, it was present year-round at depths greater than 180 m on the slope. The input of SW onto the shelf during summer resulted in more complex water mass interactions, and isotopes were able to identify the two mixed water masses, MSW and CEW. This study further confirms that during mixing of different water masses isotopes can retain their signature longer than temperature and salinity, making isotopes invaluable tracers in shelf areas with complex oceanography and mixing.

Acknowledgements. This research was supported by data collected through the Southern Australian Integrated Marine Observing System (SA1MOS). We thank the captain and crew of the R.V. Ngerin during numerous research voyages through all kinds of weather. Stable isotope analyses were done with the assistance of Kerry Klassen and April Vuletich in the Queen’s Facility for Isotope Research. Support of these analyses was from grants from the Natural Science and Engineering Research Discovery program, the Canadian Foundation for Innovation and the Ontario Innovation Foundation.

93 3.7. References

Archambeau, A-S., C. Pierre, A. Poisson and B. Schauer (1998), Distributions of oxygen and carbon stable isotopes and CFC-12 in the water masses of the Southern Ocean at 30°E from South Africa to Antarctica: results of the CIVA1 cruise, Journal of Marine Systems, 17, 25-38

Barker, P. M. (2004), The circulation and formation of water masses south of Australia and the inter-annual wind variability along the southern Australian coast, PhD thesis, University of Melbourne, Australia, pp. 351.

Bye, J. A. T. (1972), Oceanic circulation South of Australia, in Antarctic Oceanology II: The Australian-New Zealand Sector, Antarctic Research Series, vol. 19, edited by D. E. Hayes, pp. 95-100, AGU, Washington D. C.

Bye, J. A. T. (1983), Physical oceanography. In: Tyler, M.J., Twidale, C.R., Ling, J.K.. and Holmes, J.W. (Eds.), Natural History of the South East, Royal Society of South Australia.

Cirano, M. and J. F. Middleton (2004), Aspects of the mean wintertime circulation along Australia’s southern shelves: Numerical studies, Journal of Physical Oceanography, 34(3), 668-684.

Condie, S. A. and J. R. Dunn (2006), Seasonal characteristics of the surface mixed layer in the Australasian region: implications for primary production regimes and biogeography, Marine and Freshwater Research, 57, 569-590.

Corlis, N. J., H. H. Veeh, J. C. Dighton and A. L. Herczeg (2003), Mixing and evaporation processes in an inverse estuary inferred from delta H-2 and delta 0-18, Continental Shelf Research, 25(8), 835-846.

Craig, H. (1966), Isotopic composition and origin of the Rea Sea and Salton Sea geothermal brines, Science, 154(3156), 1544-1548.

Cresswell, G. R. and T. J. Golding (1980), Observations of a south-flowing current system in the southeastern Indian-Ocean, Deep-Sea Research Part I-

94 Oceanographic Research Papers, 27(6), 449-466.

Cresswell, G. R. and J. L. Peterson (1993), The Leeuwin Current south of Western Australia, Australian Journal of Marine and Freshwater Research, 44, 285- 303.

CSIRO (2012), Commonwealth Scientific and Industrial Research Organisation. Data provided by CSIRO and reproduced with the permission of CSIRO, accessed through the Data Centre: http://www.marine.csiro.au/datacentre/

Dansgaard, W. (1964), Stable isotopes in precipitation, Tellus, 16(4), 436-468.

Doubell, M. J., J. F. Middleton, D. Spencer and C. Lemckert (2014), Salt-finger enhanced mixing and vertical nutrient fluxes on Australia’s southern shelf, Continental Shelf Research, submitted manuscript, February 2014.

Epstein, S., and T. Mayeda (1953), Variation of 0-18 content of waters from natural sources, Geochim. Cosmochim. Acta, 4(5), 213-224.

Fairbanks, R. G. (1982), The Origin of Continental Shelf and Slope Water in the New York Bight and Gulf of Maine: Evidence from H2l?0/H2l60 Ratio Measurements, Journal of Geophysical Research, 87( C8), 5796-5808.

Frew, R. D., P. F. Dennis, K. J. Heywood, M. P. Meredith, and S. M. Boswell (2000), The oxygen isotope composition of water masses in the northern North Atlantic, Deep Sea Res., Part 1, 47(12), 2265 - 2286.

Gat, J. R., A. Shemesh, E. Tziperman, A. Hecht, D. Georgopoulos, and O. Basturk (1996), The stable isotope composition of waters of the eastern Mediterranean Sea, J. Geophys. Res., 101(C3), 6441-6451.

Godfrey, J. S., D. J. Vaudrey, and S. D. Hahn (1986), Observations of the shelf-edge current south of Australia Winter 1982, Journal of Physical Oceanography 16, 668-679.

Hahn, S. D. (1986), Physical structure of the waters of the Southern Australian

95 Continental shelf, Technical Report 45, Flinders Institute for Atmosphere and Marine Sciences, Flinders University of South Australia, Australia, 284pp.

Harwood, A. J. R, P. F. Dennis, A. D. Marca, G. M. Pilling and R. S. Millner (2008), The oxygen isotope composition of water masses within the North Sea, Estuarine, Coastal and Shelf Science, 78, 353-359.

Herzfeld, M. and M. Tomczak (1999), Bottom-driven upwelling generated by eastern intensification in closed and semi-closed basins with a sloping bottom, Marine and Freshwater Research, 50(7), 613-628.

Hufford, G. E., M. S. McCartney and K. A. Donohue (1997), Northern boundary currents and adjacent recirculations off southwestern Australia, Geophysical Research Letters, 24(22), 2797-2800.

James, N. R, T. D. Boreen, Y. Bone, and D. A. Feary (1994), Holocene carbonate sedimentation on the west Eucla Shelf, Great Australian Bight - A shaved shelf, Sedimentary Geology, 90(3-4), 161-177.

James, N. R, Y. Bone, L. B. Collins and T. K. Kyser (2001), Surficial sediments of the Great Australian Bight: Facies dynamics and oceanography on a vast cool- water carbonate shelf, Journal o f Sedimentary Research, 7/(4), 549-567.

James, N. P. and Y. Bone (2011), Neritic Carbonate Sediments in a Temperate Realm, Southern Australia, Springer, Berlin, 254pp.

Khatiwala, S. R, R. O. Fairbanks and R. W. Houghton (1999), Freshwater sources to the coastal ocean off northeastern North America: Evidence from H2I80/H2I(,0, Journal o f Geophysical Research, 104( C8), 18241-18255.

Lennon, G. W., D. G. Bowers, R. A. Nunes, B. D. Scott, M. Ali, J. Boyle, W. J. Cai, M. Herzfeld, G. Johansson, S. Nield, P. Petrusevics, P. Stephenson, A. A. Suskin and S. E. A. Wijffels (1987), Gravity currents and the release of salt from an inverse estuary, Nature, 527(6124), 695-697.

Lloyd, R. M. (1966), Oxygen isotope enrichment of sea water by evaporation,

96 Geochim. Cosmochim. Acta, 30(8), 801-814.

McClatchie, S., J. F. Middleton and T. M. Ward (2006), Water mass analysis and alongshore variation in upwelling intensity in the eastern Great Australian Bight, Journal o f Geophysical Research-Oceans, 111( C8), article C08007.

McConnell, M. C., R. C. Thunell, L. Lorenzoni, Y. Astor, J. D. Wright and R. Fairbanks (2009), Seasonal variability in the salinity and oxygen isotopic composition of seawater from the Cariaco Basin, Venezuela: Implications for paleosalinity reconstructions, Geochemisty, Geophysics, Geosystems, 10(6), Q06019, 15pp.

McLeay, L. J., S. J. Sorokin, R J. Rogers and T. M. Ward (2003), Benthic Protection Zone of the Great Australian Bight Marine Park: 1. Literature Review, 70 pp., South Australian Research and Development Institute (Aquatic Sciences), West Beach, South Australia.

Meredith, M. R, K. E. Grose, E. L. McDonagh, and K. J. Heywood (1999), Distribution of oxygen isotopes in the water masses of Drake Passage and the South Atlantic, J. Geophys. Res., 104(C9), 20,949-20,962.

Middleton, J. F. and M. Cirano (2002), A northern boundary current along Australia’s southern shelves: The Flinders Current, Journal of Geophysical Research- Oceans, 107(C9), article 3129.

Middleton, J. F. and G. Platov (2003), The mean summertime circulation along Australia’s southern shelves: A numerical study, Journal of Physical Oceanography, 33(11), 2270-2287.

Middleton, J. F. and J. A. T. Bye (2007), A review of the shelf-slope circulation along Australia’s southern shelves: Cape Leeuwin to Portland, Progress in Oceanography, 75(1), 1-41.

Petrusevics, P. M. (1993), SST fronts in inverse estuaries, South Australia - indicators of reduced gulf-shelf exchange, Australian Journal of Marine and Freshwater Research, 44(2), 305-323.

97 Petrusevics, P., J. A. T. Bye, V. Fahlbusch, J. Hammat, D. R. Tippins and E. van Wijk (2009), High salinity winter outflow from a mega inverse-estuary—the Great Australian Bight, Continental Shelf Research 29, 371-380.

Petrusevics, R, J. Bye, J. Luick and C. E. P. Teixeira (2011), Summer sea surface temperature fronts and elevated chlorophyll-a in the entrance to Spencer Gulf, South Australia, Continental Shelf Research, 31, 849-856.

Povinec, P. R, J. de Oliveira, E.S. Braga, J. -F. Comanducci, J. Gastaud, M. Groening, I. Levy-Palomo, U. Morgenstern and Z. Top (2008), Isotopic, trace element and nutrient characterization of coastal waters from Ubatuba inner shelf area, south-eastern Brazil, Estuarine, Coastal and Shelf Science, 76, 522- 542.

Richardson, L. E., T. K. Kyser, N. P. James and Y. Bone (2009), Analysis of hydrographic and stable isotope data to determine water masses, circulation, and mixing in the eastern Great Australian Bight, Journal o f Geophysical Research, 144, C l0016, 14pp.

Ridgway, K. R. and S. A. Condie (2004), The 5500-km-long boundary flow off western and southern Australia, Journal o f Geophysical Research-Oceans, 109(C4), article C04017.

Ridgway K. R., J. R. Dunn and J. L. Wilkin (2002), Ocean interpolation by four- dimensional least squares - Application to the waters around Australia, J. Atmos. Ocean. Tech., 19(9), 1357-1375.

SAIMOS (2012), Cruise and mooring data, Southern Australian Integrated Marine Observing System, SARDI Aquatic Sciences, West Beach, South Australia, 5022, Australia.

Turner, J. S. (1979). Buoyancy effects in fluids, pp. 368, Cambridge University Press, Cambridge.

Ward, T. M., L. J. McLeay, W. F. Dimmlich, P. J. Rogers, S. A. M. McClatchie, R. Matthews, J. Kämpf and P. D. van Ruth (2006), Pelagic ecology of a northern

98 boundary current system: effects of upwelling on the production and distribution of sardine (Sardinops sagax), anchovy (Engraulis australis) and southern bluefin tuna (Thunnus maccoyii) in the Great Australian Bight, Fisheries Oceanography, 15(3), 191-207.

Woo, M. and Pattiaratchi, C. (2008), Hydrography and water masses off the western Australian coast, Deep-Sea Research I, 55, 1090-1104.

99 Chapter 4. Upwelling characteristics and nutrient enrichment of the Kangaroo Island upwelling region, South Australia

Abstract

An analysis is presented of hydrographic and nutrient data collected over three years for the Kangaroo Island upwelling region, Lincoln Shelf, South Australia, to determine the signature of upwelled water, depth of upwelling and the source water mass being brought onto the shelf. Strong upwelling seasons were recorded during the 2007-2008 and 2009-2010 summers, while the summer of 2008-2009 had only one weak upwelling event. Strong upwelling events during February and March 2008 and February and March 2010 recorded temperatures/salinities as low as 10.4°C/34.85, and NOx/phosphate concentrations as high as 13.35/0.94 pmol/L, on the shelf. Upwelled water properties matched slope water properties between 240- 380 m, indicating water can be upwelled over depths of 200 m or more. Upwelling from these depths sources South Australian Basin Central Water of Southern Ocean origin, which is transported west along the slope by the Flinders Current. New results for nutrients show average values of NOx and phosphate during months of strong upwelling to be 6.1 times and 4.6 times greater, respectively, than during winter months, and that upwelled water can have nutrient concentrations up to 90 times higher than those in summer surface waters. This is higher than values recorded previously for the Bonney Coast. Determining nutrient enrichment during upwelling events allows for better understanding of the impact of such upwelling events on shelf ecosystems. Upwelled water was also low in silicate, a signature of Southern Ocean water masses, which has implications for phytoplankton community structure and diatom abundance on the shelf.

100 4.1. Introduction

The shelf area between the Bonney Coast and the eastern Great Australian Bight (GAB) is an important oceanographic and ecological region, where sporadic, wind- driven upwelling occurs during summer and early autumn and advects nutrients onto a predominantly oligotrophic shelf (Lewis, 1981; Schahinger, 1987; Kämpf et al., 2004; Ward et al., 2006). Such upwelling brings cool, nutrient-rich water onto the shelf and leads to high levels of primary productivity (Ward et al., 2006; Van Ruth et al., 2010a). This region supports the highest densities of sardine, anchovy and other small planktivorous fishes in Australian waters, which provide food for higher trophic level species. Eleven major commonwealth and state-run fisheries operate in this area, and all are dependent on the supply of nutrient rich water from upwelling (McLeay et al., 2003).

Two major current systems influence the continental shelf and slope in this area. The eastward flowing Leeuwin Current system extends from North West Cape, Western Australia, to southern Tasmania (Ridgway and Condie, 2004), and is an oligotrophic surface current that is strongest in winter and weakest in summer (Rochford, 1986). The South Australian Current forms part of the Leeuwin Current system, transporting Subtropical Surface Water year round from the eastern GAB to Bass Strait (Ridgway and Condie, 2004; Chapter 2 of this thesis). The Flinders Current flows below this surface current in the opposite direction west along the continental slope (Bye, 1972, 1983) and is strongest between 400-600 meters water depth (mwd) (Middleton and Cirano, 2002). Positive wind stress curl in the Southern Ocean drives equatorward Sverdrup transport that is deflected west upon reaching Australia, creating the Flinders Current (Cirano and Middleton, 2004). The Flinders Current therefore originates within the Subantarctic Zone, transporting South Australian Basin Central Water, Tasmanian Subantarctic Mode Water and Antarctic Intermediate Water north across the Subtropical Front and west along the Australian upper continental slope (Chapter 2 of this thesis; Barker, 2004; McCartney and Donohue, 2007). The interface between the Leeuwin Current System and the Flinders Current is identified to be -250 mwd, but can be as shallow as 180 mwd in the upwelling region between Kangaroo Island and the Bonney Coast during summer (Chapter 2, Chapter 3 of this thesis).

101 The Flinders Current is a northern boundary current system that has similarities to the world’s productive eastern boundary current systems off California, Peru and southern Africa (Ward et al., 2006). Several authors have suggested that the Flinders Current supplies water to the shelf during upwelling events (Ward et al., 2006; Middleton and Bye, 2007; Richardson et al., 2009). During these upwelling events, levels of primary, secondary and fish production are within the lower ranges observed during upwelling events in the eastern boundary current systems (Ward et al., 2006; Van Ruth et al., 2010a). Despite the region’s oceanographic, ecological and commercial importance and its similarities to other well-studied eastern boundary current systems, this upwelling region has not been studied extensively and several major questions remain unanswered, including the source depth of upwelling, what nutrients upwelling supplies to the shelf, how the upwelled water interacts and mixes with shelf waters, and how it varies between years. Several authors have studied the dynamics of upwelling using modelling studies and sea surface temperature (SST) data. Using SST and chlorophyll-a data, Kämpf et al. (2004) observed simultaneous upwelling events along the Bonney Coast, Kangaroo Island and the southern Eyre Peninsula in March 1998. They suggested that waters are upwelled on the narrow shelves of Kangaroo Island and the Bonney Coast, and this water is then advected west into the GAB. CTD profiles and modelling by McClatchie et al. (2006) support this theory. McClatchie et al. (2006) suggest that the nutrient-rich water observed off the southern Eyre Peninsula originates from a subsurface ‘Kangaroo Island Pool’ of upwelled water, where it is then advected towards the eastern GAB by northwesterly coastal currents. Wind-driven upwelling eventually brings this water to the surface and results in simultaneous upwelling in all three locations.

Hydrographic sections off Robe (Bonney Coast, 140°E) during upwelling events in 1983 and 1984 (Schahinger, 1987) suggest water upwelled onto the shelf seafloor comes from 250-300 mwd, and therefore water upwells over depths of 100-180 m. Kämpf et al. (2004) comment that source waters of Kangaroo Island upwelling could be coming from depths greater than 350 mwd but they do not illustrate why. In comparison, modelling by Middleton and Platov (2003) simulate upwelling occurring from 150 mwd off Kangaroo Island. Variations in the strength of upwelling between years may be the reason for these differences. Modelling of the

102 formation of the Kangaroo Island Pool of upwelled water by Kämpf (2010) suggests that submarine canyons play an important role in upwelling, bringing water from an average depth of 310 m through the canyons onto the shelf.

Only three studies have measured nutrient concentrations or considered the chemical properties of upwelled water in the region, and none have focused on the Kangaroo Island Pool on the Lincoln Shelf. On the Bonney Coast, Lewis (1981) looked at the chemical properties of water during upwelling events and found nitrate concentrations 30-70 times greater in upwelled waters compared to background levels. Richardson et al. (2009) identified water masses in the eastern GAB during an upwelling event in March 1998 and found nitrate concentrations were higher in upwelled water along the Eyre Peninsula compared to shelf waters. Van Ruth et al. (2010b) measured chlorophyll-a and macro nutrient concentrations in coastal waters off southwestern Eyre Peninsula during spring, weak upwelling and strong upwelling summers and found that macro nutrients were variable and showed no differences between seasons, despite chlorophyll-a concentrations being an order of magnitude higher during the strong upwelling summer. Identifying the chemistry of upwelled water in the Kangaroo Island Pool is important for understanding and predicting phytoplankton blooms that result from nutrient increases when upwelling occurs, in both the Lincoln Shelf and downstream eastern GAB.

Using data that span three summer seasons, this paper aims to identify the origin of upwelled water by using temperature, salinity and nutrient measurements, and to compare nutrient enrichment during upwelling events to background levels on the Lincoln Shelf. In addition, determining the depth of upwelling and the source water masses brought onto the shelf during upwelling events will clarify the influence of the Flinders Current and Southern Ocean Water on shelf ecosystems. This is the first study to focus on the nutrient concentrations of upwelled water for the Lincoln Shelf. Macro, micro and trace nutrient properties of Southern Ocean water, however, have been recorded in numerous studies (e.g. Sohrin et al., 2000; Sarmiento et al., 2004; Lai et al., 2008; Petrou et al., 2011; Hassler et al., 2012). Determining the influence of Southern Ocean Water on the Lincoln Shelf could therefore help predict what nutrients would be brought onto the shelf during future upwelling events.

103 4.2. Methods

Hydrographic data and water samples for nutrient analysis were collected aboard the RV Ngerin as part of the Southern Australian Integrated Marine Observing System (SAIMOS; SAIMOS, 2012). Research voyages were undertaken on average eight times a year between February 2008 and February 2011 on the Lincoln Shelf between Eyre Peninsula, Kangaroo Island and the mouth of Spencer Gulf (Figure 4.1). Not all CTD stations were sampled during each expedition because of weather conditions and the time of year, but CTD and water samples for nutrient analysis were collected routinely at biological stations. Collection of nutrient data did not, however, begin until August 2008. Additional data used in this study are from sampling in De Couedic Canyon (Figure 4.1) in February 2008 during Southern Surveyor voyage SS2008/02 (CSIRO, 2012).

x CTD stations AUSTRALIA Q Biological stations a SS cruise stations

• & Eyre Peninsula eastern Gulf St. GAB Spencer o Vincent Gull

Kangaroo Is. Lincoln Shelf Lacepede

de Couedic \ Canyon submarine Bonney canyons Coast 137 Longitude (°E)

Figure 4.1. Lincoln Shelf sampling locations for data analysed in this study: CTD stations sampled during SAIMOS voyages from February 2008 - February 2011, biological stations with nutrients sampled during SAIMOS voyages from August 2008 - February 2011, and CTD stations sampled during the Southern Surveyor voyage SS2008/02 during February 2008. GAB is Great Australian Bight.

104 All references to density p(S,T,p) throughout this chapter refer to potential density with a reference pressure of p = 0 m (p(S,T,0)). In-situ temperature rather than potential temperature was used to calculate potential density, therefore compression effects but not adiabatic effects were accounted for. However, for the data presented here, the errors relating to this are less than 0.3% (see Appendix B). For convenience, potential density is reported as sigma-t (a,) throughout this chapter, and is defined as:

ot = p(S,T,0)- 1000 kg/m3

The equation of state used in this chapter is UNESCO EOS-80 using practical salinity units. Pressure data is here reported as depth in meters, as 1 m is approximately equal to 1 decibar (within 1% error for the depth range 0-1200 m).

Water samples were collected for nutrient analysis at biological stations using Niskin bottles at three depths: 15 mwd, a mid-depth corresponding to the deep chlorophyll maximum (determined using CTD fluorescence), and ~10 m above the seafloor. A total of 380 samples were analysed for nutrients (see Appendix C). The majority were from water depths less than 110 m and only 16 were from depths between 120 and 400 m. 100 ml of each sample was filtered through a bonnet syringe filter (0.45 pm porosity, Micro Analytix Pty Ltd) and stored at -20°C for analysis at SARD1 Aquatic Sciences, Adelaide, Australia. Nitrate+nitrite (NOx), phosphate (P(V ) and silicate (SiCE) were determined using Flow injection analysis in a QuickChem QC8500 Automated Ion Analyser (HATCH, 2003; method for nitrate+nitrite: Lachat Quikchem method 31-107-04-1 -D, revised 16/09/2003; phosphate: Lachat Quikchem method, 31-115-01 -1 -G, revised 17/09/2003; and silicate: Lachat Quikchem method 31-114-27-1-A, revised 17/09/2003). Two replicates of each sample were analysed, and results presented as NOx in pmol/L of N, PO4 in pmol/L of P and SiCb in pmol/L. 4.3. Water Mass Context

The upwelling and nutrient characteristics in this study are herein described in the context of Lincoln Shelf water masses previously defined in Chapter 3. Hydrographic data and stable isotope values of seawater were used to identify water masses, their spatial distribution, and seasonal characteristics. Three water masses,

105 Slope Water, Subtropical Surface Water and Evaporated Water, have specific source regions and formation modes, and are present year round on the Lincoln Shelf. Two additional water masses, Mixed Slope Water and Cooled Evaporated Water, form by mixing on the shelf during summer. Water masses are briefly described below, and properties are presented in Table 4.1.

• Slope Water (SW) is a low temperature, low salinity and low stable isotope end member that is found year round at depths greater than 180 m, and is sporadically upwelled onto the shelf during summer. This water mass is sourced from South Australian Basin Central Water, which is formed at the Subtropical Front south of Australia and is transported west along the slope by the Flinders Current. • Subtropical Surface Water (STSW) is a mixed shelf water mass with intermediate salinity and stable isotope values that is transported year round into the region by the eastward Bowing South Australian Current. It is well mixed on the shelf down to depths of 180 m during winter, and is displaced at depth by the inflow of SW during summer upwelling events. • Evaporated Water (EW) is a high salinity and high stable isotope end member that is locally formed at the surface on the inner shelf and within Spencer Gulf from high evaporation and heating during summer. It outflows along the seafloor from Spencer Gulf onto the shelf during winter. • Mixed Slope Water (MSW) is a mixed shelf water mass formed by influx of SW onto the shelf during summer upwelling events. SW mixes with STSW on the shelf to form MSW with intermediate temperature and salinity, and relatively low stable isotope values. • Cooled Evaporated Water (CEW) is a mixed shelf water mass with low temperature and salinity but high stable isotope values. It is formed by vertical mixing of surface EW with low temperature and salinity bottom water during summer, and retains the high isotopic signature of EW. It is found close to the seafloor, predominantly during months with no upwelling.

106 Table 4.1: Summer water mass properties, defined by temperature, salinity and stable isotopes of seawater (from Chapter 3).

Water mass Temperature (°C) Salinity S'^O (%o) 52H (% o )

Slope Water (SW) 10.4-14.7 34.85-35.55 -0.05-0.23 1.2-2.6

Mixed Slope Water (MSW) 13.4-19.4 35.30-35.90 0.14-0.28 23-3.2

Subtropical Water (STSW) 13.3-20.1 35.35-36.20 0.18-0.47 2.6-4.2

Evaporated Water (EW) 17.5-21.3 35.70-37.30 0.35-0.87 3.5-7.4

Cooled Evaporated Water (CEW) 14.1-16.8 35.40-36.20 0.35-0.54 3.6-6.0

4.4. Results

4.4.1. Upweliing Characteristics

Weak or strong upweliing events can occur several times during the summer-autumn season, from November to April. The temperature-salinity (T-S) signature of upweliing water for the Kangaroo Island region is less than 13.5°C and less than 35.3 (Figure 4.2). This water has a linear T-S relationship. As the water upwells onto the shelf and mixes with shelf water, the linear relationship changes to a triangular relationship, spreading between warm, fresh and warm, saline end members (Figure 4.2). Cold, fresh water with temperature and salinity less than 12.5°C and 35.2 on the shelf characterised a strong upweliing event. The upwelled water mass is defined in Chapter 3 as Slope Water (SW), and can mix to temperatures and salinities of 15°C and 35.55 on the shelf while retaining a low isotopic signature (Table 4.1; Chapter 3 of this thesis).

For the 2008-2011 SAIMOS sampling period, four monthly states were identified in the region based on the season and intensity of upweliing: winter months, summer months with no upweliing, summer months with weak upweliing, and summer months with strong upweliing. Winter months were defined as May to October, with STSW widespread on the shelf. While most summer months, between November and April, had 13.5°C/35.3 SW at the shelf edge at depths >120 m, if this water did not influence the shelf at shallower depths it was considered a summer month with

107 22

Warm, saline 2 0 - summer Warm, fresh summer

18-

O o

CD

1CD 16 CL E CD

14-

Cool, saline winter jb - 12 -

Upwelled SW summer

101— 34.5 35 35.5 36 36.5 37 Salinity

Figure 4.2. Temperature vs salinity o f SAIMOS data for the Lincoln Shelf (depths 50- 150m). Four end members represent the spread in the data. Upwelled Slope Water (SW) has properties <13.5°C/<35.3. Grey lines denote density. no upwelling. The 2008-2009 summer season has very little upwelled water present. STSW was found to be widespread on the shelf during these months. When MSW was widespread on the shelf at depth, implying influx and mixing of SW, it was considered a summer month with weak upwelling. January 2010, April 2010 and February 2011 were all months with weak upwelling. Lastly, when SW with T-S of 12.5°C/35.2 was widespread on the shelf at depth, it was considered a summer month with strong upwelling. February and March 2008, and February and March

108 2010, were months with strong upwelling. T-S characteristics o f water from 100-120 mwd show the distinction between each seasonal state, especially the cool and fresh water present only during months of strong upwelling (Figure 4.3).

T-S at 100-120 mwd

* winter a summer o weak upwelling □ strong upwelling

O o

CD 3 CO CD Q. E CD

34.5 35.5 36.5 37 Salinity

Figure 4.3. Temperature vs salinity by seasonal state o f water from depths o f 100-120 m. Months o f strong upwelling have the coldest and freshest water at 100-120 mwd, while winter months have warmest and most saline water. Months o f weak upwelling are characterized by 13.5°C/35.3 water and months o f strong upwelling are characterized by 12.5°C/35.2 water. Grey lines denote density.

A ll four months with strong upwelling had SW widespread at depth across the study area. Strongest upwelling occurred during March 2010 (10.4°C/34.85 at 105 mwd) and February 2008 (10.67°C/34.9 at 90 mwd), followed by February 2010 (11.3°C/35.01 at 130 mwd) and March 2008 (12.38°C/35.16 at 124 mwd) (Table

109 4.2). An RV Southern Surveyor voyage a few days after the February 2008 SAIMOS voyage also sampled very low temperature and salinity water, of 10.3°C/34.85 at 100 mwd, in the same region. These temperatures and salinities are considerably less than those present during non-upwelling months, such as during February and March 2009 (Figure 4.4). Temperature and salinity at the base of the seasonal thermocline for strong upwelling months was less than 12.5°C/35.15, while it was greater than 14°C/>35.5 during non-upwelling months. Differences are also seen in off-shelf profiles down to 350 mwd (Figure 4.4).

Table 4.2. Temperature/Salinity (T/S) properties and corresponding depths of strongest upwelling recorded during SAIMOS voyages. The first column presents the lowest T/S measured at the bottom of the seasonal thermocline during SAIMOS voyages, the second column presents the lowest T/S recorded on the shelf during SAIMOS voyages, labelled (A), and the third column presents the corresponding long-term average summer depth of (A) based on historic data. Historic data are from 136°E, collected during past research voyages and accessed through the CSIRO databases (CSIRO, 2012). *from SS2008/02 research voyage, 136.5°E, 17th-21st February 2008.

M onth Lowest T/S at bottom of Lowest T/S recorded on Summer average depth

seasonal thermocline the s h e lf (A ) at 136°E (historic) of

(A)

February 2008 11.5°C/35.0 at 65 mwd 10.67°C/34.9 at 90 mwd 3 6 0 m w d

10.3°C/34.85 at 100 mwd* 3 8 0 m w d

March 2008 12.5°C/35.15 at 50 mwd 12.38°C/35.15 at 124 mwd 240 mwd

February 2010 12.4°C/35.15 at 55 mwd 11.3°C/35.0 at 130 mwd 315 mwd

March 2010 11.13°C/35.0 at 55 mwd 10.4°C/34.85 at 105 mwd 380 mwd

Off-shelf profiles were highly variable both within and between months, depending on the location along the shelf and to what extent the thermocline rose during upwelling events. Off-shelf profiles in the east of the study area consistently showed shallower isotherms, indicating stronger upwelling. In addition, historic data from past research voyages (CSIRO, 2012; not shown) were examined and indicate that isotherms further east along the Bonney Coast (~140°E), are on average raised by

110 60-75 m more than those of Kangaroo Island (~136°E). This has also been observed in earlier studies (Middleton et al., 2007; Middleton and Bye, 2007).

(a) February-March temperature profiles (b) February-March salinity profiles

Temperature (°C) Salinity

Figure 4.4. Profiles o f temperature (a) and salinity (b) comparing upwelling months February and March 2008 and 2010 (blue), with non-upwelling months February and March 2009 (red).

2008 Upwelling Season

SAIMOS sampling for this study began in February 2008. SAIMOS voyages sampled upwelled water from the 12th-14th February 2008 and again from the 15th- 17th March 2008. Sea surface temperature (SST) images show that several strong upwelling events occurred between the Eyre Peninsula (135°E) and the Bonney Coast (140°E) from January to March. The upwelling event sampled on the February voyage began on the 30th January, with cold, 15°C surface water along the Bonney Coast, with a small surface signal o ff the western tip of Kangaroo Island and o ff western Eyre Peninsula. By the 5th February the Bonney Coast cold plume had spread to the south of Kangaroo Island, but the Kangaroo Island signal remained small. On the 13th and 14th February this signal grew to its largest extent (Figure 4.5). This upwelling peak encompassed all three upwelling centres - the Bonney Coast upwelling had spread as far as southwest o f Kangaroo Island, and upwelled water was present along western Eyre Peninsula. The timing o f the February voyage captured this event.

I l l 200 1000m Altimetric sealevel 14-Feb-2008 (0.1m contours) and velocity for 04:06Z 14-Feb 04Z 0 5m/e(1kt 24h)

Adelaide -3 5 - ANMN 4h avg < , 10-30m < — ,50-70m 4 —*80-120m % , 150-200m O Argo drifter$@12h to > 14-Feb 06 Z

Figure 4.5. Sea Surface Temperature (SST) image of a strong upwelling event on the 14th February 2008. Consecutive upwelling occurs along the Bonney Coast (east), the western tip of Kangaroo Island and along the western Eyre Peninsula (west). Upwelling along the Bonney Coast is most predominant and spreads towards Kangaroo Island. Source: IMOS OceanCurrent (IMOS, 2012). NI indicates the location of Neptune Island weather station, referred to in Section 4.5.2.

By the 17th February the Bonney Coast plume had receded to the east of Kangaroo Island but still remained large across the Lacepede Shelf. Another upwelling peak occurred across all three upwelling centres on the 21st February and again on the 29th February. By the 5th March only a thin band of cold water along the Bonney Coast was present. During the March voyage, no surface signal was present off Kangaroo Island or the Eyre Peninsula, and only a small area of cold water was present along the Bonney Coast.

During the February 2008 voyage, coldest and freshest water sampled at the bottom of the seasonal thermocline at ~60m was ~11.5°C and 35.0. This water was not present at the shelf edge but close to the western tip of Kangaroo Island. Bottom

112 contours of temperature and salinity illustrate the location of this cold water, with warmer and more saline water closer to the shelf edge (Figure 4.6a,b). This upwelling plume appears to be coming from the east, in agreement with the SST images that show the Bonney Coast upwelling plume spreading towards Kangaroo Island (Figure 4.5).

Figure 4.6. Bottom contours of temperature and salinity for months of strong upwelling during 2008. a) and b) February 2008, c) and d) March 2008.

The source of upwelled water onto the shelf is not as clear in bottom temperature and salinity contours from March 2008 (Figure 4.6b,c). Warmer and more saline water adjacent to the de Couedic Canyon suggests that upwelling was not coming up through this major submarine canyon. The upwelled plume was still, however, situated off southwestern Kangaroo Island. The coldest and freshest shelf water sampled during March, of 12.38°C/35.15 at 124 mwd was not present in the off-shelf profile sampled to 380 mwd. The off-shelf profile did not have the raised isotherms

113 that are observed during an upwelling event, suggesting that the upwelling event had passed. Therefore, it is likely that shelf water measured on the March voyage was upwelled at an earlier time.

2010 Upwelling Season

The strong upwelling season of 2010 started in November 2009. There was a drop of 1.5°C in temperature at 100-120 mwd between October and November 2009, and average salinities were as low as 35.3 in November 2009, signalling the onset of the upwelling season. By comparison, average salinities at 100-120 mwd in November and December 2008 were greater than 35.8, before salinity dropped to only 35.5 in January-March 2009. Therefore, for the strong upwelling season of 2009-2010, upwelling began in November and strengthened in February and March. Bottom water properties measured in November may predict whether a strong or weak upwelling season is expected. Bottom salinity contours during November 2008 and 2009 highlight the differences between the two years (Figure 4.7). Winter conditions were still present in November 2008 with saline outflow from Spencer Gulf still visible. On the other hand November 2009 was associated with summer conditions with cool, fresh water present. These observations however are based on only one weak and one strong upwelling season; more data are needed to support this hypothesis and support the conclusion that November consistently marks onset and predicts the strength of the coming upwelling season. Anecdotal evidence based on the presence of Blue Whale sightings, however, suggests that upwelling along the Bonney Coast also begins in November (Gill, 2002).

SST images (e.g. Figure 4.8) show consistently cold (~15°C) water along the Bonney Coast during February and March 2010. Cold surface water was also present along the western Eyre Peninsula between the 28th February and 19th March. A surface signal off the western tip of Kangaroo Island was not as prevalent however, with 15-16°C water only present during 3-5th March and the 12-13th March. The Bonney Coast upwelling was at its greatest spatial extent during these times, spreading along the shelf edge south of Kangaroo Island (Figure 4.8). The strong 12- 13th March upwelling event was sampled during the 16- 18th March voyage, recording the lowest temperatures and salinities on the shelf during the study period, of 10.4°C/34.85 at 105 mwd.

114 November 2008 - Salinity November 2009 - Salinity

35°S

36°S

20' 40' 136°E 20' 40'

Figure 4.7. Bottom salinity contours for a) November 2008 and b) November 2009. November 2008 precedes a season with minimal upwelling, and still exhibits outflow from Spencer Gulf. In comparison, November 2009 precedes a season with strong upwelling and has low salinity ocean water on the shelf.

Extensive sampling during the 16-18th February 2010 voyage resulted in good spatial coverage of data on the shelf. Bottom temperature and salinity contours show cooler and less saline water as a pool on the shelf west of Kangaroo Island, with a path to the shelf edge at around 136°20’E (Figure 4.9a,b). This water appears to be coming directly from the shelf edge, possibly from de Couedic Canyon to the south of Kangaroo Island. The cold pool is situated in the center of the Lincoln Shelf, to the west of Kangaroo Island. Sampling close to the shelf edge during the 16-18lh March 2010 voyage was limited and focused around de Couedic Canyon, so it is not possible to rule out upwelling sourced from the shelf edge east or west of the canyon (Figure 4.9c,d). Based on the SST images, however, upwelled water may again be coming from the Bonney Coast, which clearly spreads west towards Kangaroo Island (Figure 4.8).

During the March 2010 voyage, very cool and fresh water, less than 12°C and 35.0, was present in depth ranges as shallow as 40-60 mwd. The coldest and freshest water recorded on the shelf during this month (10.4°C/34.85), however, was not detected in the March 2010 slope profile to 365 mwd. Like during March 2008, the off-shelf profile suggests that the upwelling event had passed. Water with these characteristics was detected in two profiles sampled during February 2010, at 300 and 306 mwd. This deep water sampled in February 2010 could be the same water found on the shelf during March 2010. Water upwelled during February along the Bonney Coast

115 could also have been transported west and upwelled in the Kangaroo Island upwelling pool during March. This situation would indicate that the strong upwelling event during March 2010 was sourced from -300 mwd. The average depth of this water in historic slope profiles (Table 4.2) is 380 mwd, therefore it appears that isotherms can be raised significantly during upwelling events, and upwelled water is sourced from substantial depths. Mechanisms include wind forced upwelling, canyon upwelling and upwelling of the thermocline due to El Nino summers (Middleton et al„ 2007). 11:50:20 Hobarl 11:50:20

134 135 136 137 138 139 140 141

Figure 4.8. Sea Surface Temperature (SST) image of a strong upwelling event on the 3rd March 2010, showing consecutive upwelling along the Bonney Coast (east), the western tip of Kangaroo Island and along the western Eyre Peninsula (west). Source: IMOS OceanCurrent (IMOS, 2012).

116 (C) March 2010 - Temperature (°C) ( d ) March 2010 - Salinity

Figure 4.9. Bottom contours of temperature and salinity for months of strong upwelling during 2010. a) and b) February 2010, c) and d) March 2010.

4.4.2 Nutrients and Upwelling

Identifying nutrient characteristics by season and upwelling state is useful to determine average shelf nutrient values during upwelling events, for future predictions of nutrient inputs to the shelf. To this end, oxides of nitrogen

(nitrate+nitrite, NOx), phosphate (PO4) and silicate (SiÜ2 ) were routinely measured from samples at three depths between August 2008 and February 2011. Average shelf values of nutrients for each season are presented in Table 4.3. Si* is the relative abundance of silicate to nitrate, and has been used as a conservative tracer of

Southern Ocean water masses (Sarmiento et al., 2004). In this case Si* = [SiÜ2 - NOx], measured in pmol/L. Sarmiento et al. (2004) used nitrate instead of NOx, however it is herein proposed that NOx is an acceptable substitution because nitrite is a small fraction ofNOx in shelf waters (global average 1.4%) and is insignificant in water below the photic zone (global average 0.02%) (Gruber, 2008). Unfortunately

117 no nutrient data are available for the strong upwelling events of February and March 2008, therefore nutrient characteristics of strong upwelling events presented here are from data collected during February and March 2010. Samples from April 2010, while not having water less than 13.5°C and 35.3, recorded high nutrient concentrations and low stable isotope values, possibly due to remnant upwelled water from March 2010, and is included in this section as a strong upwelling month.

Table 4.3. Average values of nutrients by seasonal state (in pmol/L) for the Lincoln Shelf (depths less than 150 mwd). Total range of values in parentheses. March 2009, January 2010 and February 2011 were weak upwelling months; February, March and April 2010 were strong upwelling months (no data for February and March 2008).

Nutrient Winter months Summer months, Weak upwelling Strong upwelling

(May-Oct) no upwelling months (pmol/L) months (pmol/L)

(pmol/L) (Nov-April)

(pmol/L)

NO* o f N 0.50 (0-2.42) 0.74 (0-3.37) 1.46 (0-5.8) 3.19(0-13.35)

P 0 4 o fP 0.05 (0-0.19) 0.06 (0-0.25) 0.13 (0-0.39) 0.23 (0-0.94)

SiO, 0.86 (0.33-1.65) 0.71 (0-2.25) 0.88 (0.14-2.0) 0.79 (0.08-3.45)

Si* 0.35 (-1.08-1.21) -0.02 (-2.2-1.03) -0.58 (-4.38-1.08) -2.44 (-9.9-0.28)

Nutrient relationships

NOx and phosphate data show a positive linear relationship for the Lincoln Shelf (Figure 4.10), with a slope fit of y = 12.72x - 0.08 and R2 = 0.83. The average NOx:phosphate ratio was 12.7:1, which is slightly lower than the Redfield ratio of 16:1. This ratio increased to 14:1 during strong upwelling months. These strong upwelling months exhibited robust linear relationships between NOx and phosphate; the relationship in March 2010 had an R2 of 0.99, whereas February 2010 and April 2010 had R2 values of 0.95 and 0.91, respectively. There was almost no relationship between the two nutrients during wintertime, with R2 values around 0.12.

118 (a) N0X vs phosphate

z 6

* winter a summer o weak upwelling □ strong upwelling

0.4 0.6 Phosphate (pmol/L)

(b) NOx vs silicate

Silicate (pmol/L)

Figure 4.10. Nutrient relationships by seasonal state for a) NOx and phosphate and b) NOx and silicate. NOx vs phosphate shows a good linear relationship but NOx vs silicate has more variation without a clear relationship.

119 The highest values for NOx and phosphate were 13.35 pmol/L and 0.94 pmol/L respectively, recorded at 105 mwd close to the shelf edge southwest of Kangaroo Island during March 2010. To place these results in a regional context, such high nutrient values are present at -400 mwd in annually averaged data for 140°E (refer to Figure 2.6 in Chapter 2). This depth is in agreement with the upwelling source depth of 380 m estimated by temperature and salinity data (Table 4.2). During February and April 2010 values were as high as 9.1 and 8.2 pmol/L for NOx and 0.56 and 0.51 pmol/L for phosphate, respectively. In comparison, typical values recorded during summer months with no upwelling were less than 3.5 pmol/L NOx and less than 0.3 pmol/L phosphate, and typical values for winter months were less than 1 pmol/L NOx and less than 0.15 pmol/L phosphate (Figure 4.10a).

The NOx and silicate relationship was more complicated than the NOx and phosphate relationship (Figure 4.10b). Silicate levels remained generally low (less than 2 pmol/L) and did not change significantly between months (Table 4.3). Only three samples with very high NOx (greater than 8 pmol/L), recorded during March 2010, had silicate levels higher than data recorded during other months (Figure 4.10b). Significant variations in the NOx:silicate ratio were instead a result of dramatic increases in NOx during upwelling months. NOx:silicate ratios were high during the strong upwelling months of February and March 2010, with ratios of 4.7:1 and 2.8:1, respectively. In comparison, a ratio of — 1:1 was present for the rest of the data. Like the NOx:phosphate relationship, upwelling months had stronger linear NOx:silicate relationships than winter months. The high NOx to relatively low silicate signal during upwelling months resulted in negative Si* values (Figure 4.11). Si* values were as low as -7.9 pmol/L and -9.9 pmol/L in February 2010 and March 2010, respectively, indicating that NOx was increasing significantly more than silicate during strong upwelling months.

Seasonal states

To compare the degree that shelf nutrients increase during upwelling events, nutrient levels for both winter and summer with no upwelling are here considered as background levels (Table 4.3). NOx and phosphate increased significantly on the shelf during upwelling months; average shelf values during strong upwelling months were 6.1 times higher in NOx and 4.6 times higher in phosphate than winter months.

120 When comparing the upwelled water mass (SW), average values of NOx were 7.7 times greater in SW than background summer shelf values (STSW), 10.6 times greater than background winter shelf values (STSW), and 36 times greater than the nutrient depleted summer surface values (EW). The most nutrient-rich water upwelled on the shelf in March 2010 had NOx values 18 times higher and 26 times higher than summer and winter STSW background levels, respectively, and 90 times higher than summer surface waters (EW).

Si* profiles by seasonal state

®0 A m

• winter

o weak upwelling ° strong upwelling

Si* (pmol/L)

Figure 4.11. Si* profiles by seasonal state. Si* = [SiC>2 - NOx], measured in gmol/L. Months o f weak and strong upwelling show very negative values, compared to winter months which have predominantly positive values. Dotted line is Si* = 0 pmol/L.

Compared to NOx and phosphate, silicate did not show any trend between states (Table 4.3), signifying that upwelled water is relatively low in silicate and upwelling events do not result in increased shelf values, except for three samples very high in NOx sampled during March 2010 (Figure 4.10b). Silicate concentrations during winter months rarely dropped below 0.5 pmol/L, despite waters being low in NOx and phosphate (Figure 4.10b). Therefore, low NOx and silicate values meant that winter months had the highest Si* values of all the seasonal states, with a positive average value (Table 4.3). By comparison, strong upwelling months had low silicate values accompanied by significant increases in NOx, resulting in Si* values as low as -9.9 pmol/L. Si* profiles highlight the very negative values during months of strong

121 upwelling, and the predominantly positive values during winter months (Figure 4.11).

Nutrient levels peak at ~ 100 mwd (Figure 4.12) for all summer seasonal states, which is the result of upwelled nutrient-rich SW that lies close to the shelf edge. Elevated nutrient levels in the upper 60 mwd are only observed during months of strong upwelling, however, suggesting that the nutrients supplied during weak upwelling events are very quickly utilized. While timing of an upwelling event could have an effect on what nutrients were measured, timing of SAIMOS voyages when sampling weak upwelling events was not different relative to when sampling strong upwelling events. The photic zone can be -100 mwd in the eastern GAB (Van Ruth et al., 2010a), therefore even during weak upwelling events, nutrient rich water would be available to autotrophs in the photic zone. Weak upwelling events do not show a surface upwelling signal, however. Thus it is likely possible to observe increases in phytoplankton without having any surface evidence o f upwelled water. This is in agreement with conclusions made by Van Ruth et al. (2010a,b), who found that high levels o f primary productivity occurred even if upwelling did not reach the surface.

(a) NO, profiles (t>) Phosphate profiles (c) Silicate profiles

« winter

o weak upwelling □ strong upwelling

5 10 0.2 0.4 0.6 0.8 NO„ (pmol/L) Phosphate (pmol/L) Silicate (pmol/L)

Figure 4.12. Depth profiles o f a) NOx, b) phosphate and c) silicate presented by seasonal state. Elevated NOx and phosphate concentrations are evident during strong and weak upwelling months, but there is a lot o f overlap in silicate concentrations between states.

122 Upwelled water at 100-110 mwd during March 2010 had significantly higher nutrient levels than waters recorded in the slope profile measured to 365 mwd (Figure 4.12). At 365 mwd, NOx and phosphate were 7.6 pmol/L and 0.6 gmol/L respectively, compared to 9.9-13.35 pmol/L for NOx and 0.7-0.94 pmol/L for phosphate at 100-110 mwd. To supply such high nutrient values to the shelf, water is either coming from the Bonney Coast upwelling region or through submarine canyons, where nutrient profiles could be different, or the nutrient contours are no longer raised due to upwelling, as shown by temperature and salinity data. Based on summer averaged historic data (CSIRO, 2012), values of 13.35 pmol/L and 0.94 pmol/L for NOx and phosphate, respectively, match nutrient values from at least 350 mwd in both regions (refer to Figure 2.6 in Chapter 2).

4.5. Discussion

4.5.1 Depth of upwelling

Temperature and salinity of waters upwelled onto the Lincoln Shelf during strong upwelling events match the characteristics of off-shelf waters from depths greater than 240 m, based on summer averaged historic data. For two of the four strong upwelling events, February 2008 and March 2010, the average depth of upwelled water found in slope profiles was 380 m (Table 4.2). In addition, the high nutrient values present on the shelf during March 2010 match nutrient values from at least 350 mwd, which matches well with the summer averaged data for temperature and salinity. For these strong upwelling events, water appears to be upwelled over depths of 200 m or more.

The cold, low salinity, high nutrient water present on the shelf during strong upwelling events was not always found in off-shelf profiles sampled to 365 mwd. This situation occurred during both March 2008 and March 2010. It is hypothesised that the slope isotherms were no longer raised due to upwelling this late in the summer season, and had instead returned to average depths present during the rest of the year. Upwelled water on the shelf, therefore, was left over from an earlier upwelling event, or was being transported west from the Bonney Coast upwelling. This supports the presence of a Kangaroo Island Pool of upwelled water first

123 identified by McClatchie et al. (2006), where previously upwelled water sits at depth and is then upwelled to the surface during the next upwelling event.

Isotherms in off-shelf profiles were raised during February 2010 compared to March 2010, and water at 300 mwd during February matched shelf water at 105 mwd during March. It is possible that this water at 300 mwd was then upwelled onto the shelf in March, or upwelled from this depth along the Bonney Coast and transported west to Kangaroo Island. For these samples, however, lower nutrient values in the 300 mwd samples, and differences in the NOx:silicate ratio between the shelf and slope samples suggests it is not the same water. Despite this, it is plausible that isotherms are raised from an average summer depth of 380 m to 300 m during an upwelling event, and differences in nutrients are a result of the timing and location of collection for the samples that are being compared.

SST images in March 2010 (e.g. Figure 4.8) show a spread of water from the east reaching south of Kangaroo Island. Bottom contours of both temperature and salinity during February 2008 also suggest an eastern source of upwelled water (Figure 4.6a,b). Deep profiles from historic data stored by CSIRO (CSIRO, 2012) for the Kangaroo Island and Bonney Coast regions suggest that during summer the Bonney Coast has higher nitrate contents at any given depth than Kangaroo Island. There is however, overlap between the two regions and there is also a greater density of data for the Bonney Coast. SAIMOS samples by contrast show that the high NOx recorded during March 2010 matches with Bonney Coast deep water at -300 mwd or deeper, therefore, regardless of where the upwelling is coming from, characteristics match slope water at a minimum of 300 mwd. Comparing density and Si* profiles from strong upwelling months with historic data of the Kangaroo Island region also confirms this depth. Combining the evidence from off-shelf profiles, nutrient data and summer averaged depths from historic data, water is being upwelled from depths between 300 and 380 m during February 2008 and March 2010. Previous literature proposes various source depths of upwelling water, likely due to variations in strength of different upwelling events. Schahinger (1987) suggests 250-300 mwd, based on hydrographic sections for the Bonney Coast, which matches well with the depths proposed herein. Modelling of upwelling through submarine canyons by Kämpf (2010), suggests and average of 310 mwd, which again matches well with depths determined in this study. 124 4.5.2 Upwelling mechanisms and variation between years

The 2008 and 2010 upwelling seasons were very similar. Lowest temperatures and salinities recorded during both seasons were comparable and matched slope water properties at 380 mwd (Table 4.2), while SST images showed almost identical upwelling plume development and spread from the Bonney Coast west to Kangaroo Island (e.g. Figures 4.5, 4.8). The 2009 summer season, however, lacked any major upwelling events and had significantly warmer and more saline water on the shelf (Figure 4.4).

Upwelling intensity is dependent on both the strength of upwelling favourable winds and the presence of El Nino events in the eastern Pacific (Middleton et al., 2007). Southeasterly winds are upwelling favourable during summer, as they drive a westward shelf current, induce transport of water offshore, and promote upwelling. In addition, El Nino events influence the southern continental shelf and slope by lowering sea level and raising the thermocline (Middleton et al., 2007). During an El Nino year, downwelling is reduced in winter and upwelling is enhanced the following summer. Wind stress and El Nino Southern Oscillation (ENSO) events appear to be independent of each other, as Middleton et al. (2007) found little correlation between the two proxies. Instead, ENSO signals are propagated along the wave-guide of the shelf slope from the western Pacific, via Western Australia, rather than through meteorological effects (Li and Clarke, 2004; Middleton et al., 2007).

Averaged summer and winter wind stress data for 2008-2010 shows that the 2008 summer had the most upwelling favourable conditions (Figure 4.13). Weekly averaged data, however, shows significant variation (Figure 4.14). An early summer peak in wind stress is observed during both November 2007 and 2009, but is absent during November 2008. The 2009-2010 summer season had variable winds, with a drop in wind stress over December and January, and an increase in February. A very large peak in wind stress during March 2010 correlates with the strongest upwelling event recorded during this study.

La Nina conditions were present during the 2007-2008 and 2008-2009 summer seasons, therefore above average wind stress may have been the dominant factor in the February and March 2008 strong upwelling events. Middleton et al. (2007), however, found that significant upwelling occurred in La Nina summers that 125 followed strong El Nino events. La Nina summers of 1984 and 1999 followed the strong El Nino summers o f 1983 and 1998, and had anomalously cold water on the shelf. The authors attributed this to either a) weaker wind-forced downwelling in the winter following the El Nino event, resulting in remnant cold water upwelled from the previous year being brought back onto the shelf or b) higher than average upwelling favourable winds. The 2007-2008 summer followed an El Nino event in 2006; therefore it is possible that the strong upwelling event during February 2008 was a combination of these two mechanisms.

0.1

0.05

«■ t 0 CO(/> CD tn ~o I - 0.05

- 0.1

- 0.15 summer winter summer winter summer winter 2008 2008 2009 2009 2010 2010

Figure 4.13. Mean wind stress for Neptune Island, South Australia (for location see Figure 4.5), calculated for summer (January-March) and winter (June-September). Positive values indicate upwelling favourable conditions and negative values indicate downwelling favourable conditions (SE-NW grid). Dashed lines represent the long-term summer and winter means.

Lower than average wind stress in association with La Nina conditions resulted in little to no upwelling during early 2009. An El Nino event began in mid 2009, which would have affected the 2010 summer upwelling season. The occurrence of strong upwelling favourable winds during March 2010 resulted in upwelling that was likely enhanced by the El Nino event. Middleton et al. (2007) found that in situ temperature measurements, but not necessarily SSTs, were much colder during El Nino summers. Measured temperatures of 10.5-11.5°C at 80-120 mwd during the El 126 Nino summers 1998 and 2003 were found to be 2-3 standard deviations colder than the mean. These temperatures are comparable to temperatures recorded in this study during the 2008 and 2010 summer upwelling events.

Figure 4.14. Weekly averaged wind stress for Neptune island for the period January 2007- December 2010. Positive values represent upwelling favourable conditions. Dashed lines indicate strong upwelling events February and March 2008 (A and B) and February and March 2010 (C and D).

It is likely that a combination of strong upwelling favourable winds and the influence of El Nino signals caused strong upwelling to occur during February 2008 and March 2010, as upwelling characteristics during these two months were remarkably similar. Upwelling favourable winds were below average and variable during January and February 2010; therefore El Nino conditions may have played a role in the persistence of strong upwelling during February and March 2010.

4.5.3 Nutrient enrichment during upwelling events

Slope Water (SW) brought onto the shelf during strong upwelling events had the highest values of NOx, phosphate and silicate of any water mass. Highest NOx and phosphate values were 13.4 pmol/F and 0.94 gmol/F, respectively, measured at 105 mwd southwest of Kangaroo Island in March 2010. In comparison, winter values

127 were typically less than 1 pmol/L NOx and less than 0.15 pmol/L phosphate, with average NOx of 0.5 pmol/L and average phosphate of 0.05 pmol/L. These wintertime results provide a nutrient signature for Subtropical Surface Water (STSW) transported by the South Australian Current. This water flows east along the southern Australian margin within the top 200 mwd, and is well mixed from surface to seafloor. Any nutrients are quickly taken up by phytoplankton in the photic zone, resulting in very low nutrient levels. The surface waters in this region have been labelled as oligotrophic overall (Longhurst, 1998), therefore upwelling events episodically bring significant nutrients to support shelf ecosystems and fisheries.

Average values of NOx in SW were 10.6 times greater than STSW during winter, 7.7 times greater than STSW during summer, and 36 times greater than the nutrient depleted surface EW during summer. In addition, average shelf values during months of strong upwelling were 6.1 times higher in NOx and 4.6 times higher in phosphate than winter months. This is the first study, to our knowledge, that has presented data on significant nutrient enrichment of the Kangaroo Island Pool of upwelled water. High nutrient concentrations have been recorded for the Bonney Coast upwelling system to the east of Kangaroo Island (Lewis, 1981), with nitrate concentrations 30-70 times greater in upwelled water compared to background levels of 0.1-0.9 mmol/m’ (equivalent to pmol/L). These background levels match well with winter values recorded during this study. Highest values of nitrate were 7 pmol/L (Lewis, 1981), whereas in this study recorded NOx values were almost twice that value, reaching as high as 13.4 pmol/L in March 2010. Nitrate alone was not measured in this study and so cannot be compared, but it is likely that nitrite values in deep water are not high enough to make a significant difference and NOx is comparable to nitrate.

The most nutrient-rich upwelled water was up to 26 times higher in NOx than average winter background values of 0.5 pmol/L, and up to 90 times higher in NOx than average summer surface water values of 0.15 pmol/L. This is a significant enrichment and the highest recorded to date. To the west of the Lincoln Shelf, in the eastern GAB, Richardson et al. (2009) found nitrate concentrations were higher in upwelled water along the Eyre Peninsula, with nitrate concentrations of 5 pmol/L equivalent to concentrations present at ~250m depth. This concentration is still much lower than the 13.4 pmol/L of NOx measured southwest of Kangaroo Island during 128 this study. Analysis of productivity levels between upwelling and downwelling events off southwestern Eyre Peninsula by van Ruth et al. (2010b) showed highest productivity during upwelling events, but no major variations in macro-nutrients between upwelling and downwelling events. Highest values of NOx were ~7 pmol/L, again much less than concentrations recorded in this study. Van Ruth et al. (2010b) hypothesised that upwelling brings micro-nutrients that are being utilized by phytoplankton. Based on results of this study, upwelled water brings considerable macro-nutrients to the shelf to promote primary productivity in the region.

4.5.4 Silicate signature and source of upwelled water

The ratio of NOx to silicate appears to be a good indicator of SW, with a high NOx:silicate ratio and relatively low Si* values during months of strong upwelling (e.g. Figure 4.11). This signature is also retained as SW mixes to MSW. By comparison, during winter months and summer months with no upwelling, shelf waters had low NOx values, less than 4 gmol/L, but had silicate values as high as 2.3 pmol/L (Figure 4.10b), which were comparable to the highest shelf values observed throughout the year.

SW on the Lincoln Shelf had low Si* values due to NOx enrichment, which reached a minimum on the shelf o f-9.9 pmol/L in March 2010 (Figure 4.11). This is within South Australian Basin Central Water (SABCW) values and very close to Tasmanian Subantarctic Mode Water (TSAMW) values o f-10 pmol/L (Chapter 2 of this thesis; Sarmiento et al., 2004). The Southern Ocean has high NOx:silicate ratios and low Si* values, because iron limitation increases silicate uptake by diatoms and results in a relative abundance of nitrate to silicate in surface waters (Franck et al., 2000). The Si* proxy has been utilized as a conservative tracer of Southern Ocean water masses through the southern hemisphere and into the north Pacific (Sarmiento et al., 2004; Bibby and Moore, 2011), as Si* values in Southern Ocean surface waters are the lowest in the global ocean. The low Si* of upwelled water confirms the influence of Southern Ocean water masses on the shelf, and has implications for diatom production in shelf ecosystems, as Si* is a significant control on diatom community structure (Bibby and Moore, 2011).

In comparison to Southern Ocean water, the Leeuwin Current has a high silicate signature on the west coast of Australia (Lourey et al., 2006). The westward flow of 129 the Leeuwin Current and South Australian Current along the southern margin of Australia results in a decrease in silicate enrichment from west to east (Chapter 2 of this thesis). STSW carried by the South Australian Current, however, is still enriched in silicate compared to SW on the Lincoln Shelf. This highlights the different source regions of the two water masses - STSW sourced from the west coast of Australia and SW sourced from the Southern Ocean.

Both the ratio of NOx to silicate and Si* values, therefore, are distinctly different for Southern Ocean water and Leeuwin Current water, providing an insight into the influence of both these source waters on the Kangaroo Island region. SABCW is identified along the southern margin of Australia as shallow as 180 mwd in the upwelling region between the Bonney Coast and Kangaroo Island (Chapter 2 of this thesis). SABCW forms at the Subtropical Front south of Australia, and is transported north and then west along the southern Australian continental slope by the Flinders Current. It has temperature and salinity of 10-12°C and 34.8-35.1, respectively, and nitrate concentrations of 5-15 pmol/L. Upwelled water on the Lincoln shelf during strong upwelling events matches these SABCW properties. The Flinders Current, therefore, sources this SABCW from the Southern Ocean, which is then upwelled onto the shelf in the upwelling region. The Flinders Current has been proposed as the source of upwelling in the region (Ward et al., 2006; Richardson et al., 2009). This is, however, the first paper to match the specific hydrographic and nutrient properties of SABCW from the Southern Ocean with upwelled water. TSAMW on the slope is characterized by a layer of relatively constant temperature and density between ~400 and 650 mwd (Chapter 2 of this thesis), and therefore upwelling probably does not come from deep enough to directly source this water mass. The coldest and freshest March 2010 upwelled water is, however, very close in attributes to the upper boundary of TSAMW; T-S properties of 10°C, 34.8 and Si* values less than -10 pmol/L of TSAMW are very close to 10.4°C, 34.85 and -9.9 pmol/L of upwelled water, highlighting the strong association with TSAMW to upwelling in the Kangaroo Island region. Understanding major and trace nutrient properties of Southern Ocean water masses, which have received much more attention than nutrient studies in the Kangaroo Island region (e.g. Sohrin et al., 2000; Sarmiento et al., 2004; Lai et al., 2008; Petrou et al., 2011; Hassler et al., 2012), will allow for nutrient concentrations to be estimated during future upwelling events if temperature

130 and salinity are the only data available.

4.6. Conclusions

Analysis of four strong upwelling events and several weak upwelling events between February 2008 and February 2011 has provided insight into hydrographic and nutrient properties of upwelled water on the Lincoln Shelf. The upwelling signature is less than 13.5°C and 35.3 during weak upwelling events and less than 12.5°C and 35.2 during strong upwelling events. Strong upwelling events occurred during February and March 2008 and February and March 2010. The coldest and freshest water recorded on the shelf during the study period had temperatures and salinities of 10.4°C and 34.85, respectively, sampled at 100-110 mwd during February 2008 and March 2010. Temperature, salinity and nutrient data from this study suggest that at least 300 mwd is needed to supply such cold, fresh and nutrient-rich water to the shelf, indicating water is upwelled over depths of 200 m or more. The average depth of this water recorded in slope profiles from historic data is 380 mwd, therefore isotherms can be raised significantly during upwelling events.

Entry of upwelled water onto the shelf remains uncertain during some upwelling months due to the spatial distribution of data, but appeared to be coming from the Bonney Coast during February 2008 and from the shelf edge south of Kangaroo Island during February 2010. SST images during March 2010 suggest that upwelled water from the Bonney Coast spreads west along the shelf edge south of Kangaroo Island, and could have sourced upwelled water on the Lincoln Shelf during this time. If this is the case, the three upwelling centres along the South Australian coast - the Bonney Coast, Kangaroo Island, and western Eyre Peninsula - could all be sourced from the Bonney Coast.

Upwelled water supplies significant macro-nutrients to an otherwise oligotrophic shelf. Water upwelled onto the shelf during strong upwelling events had the highest values of NOx, phosphate and silicate. The most nutrient-rich water on the shelf in March 2010 had NOx and phosphate concentrations of 13.35 and 0.94 pmol/L, respectively. These values were 18 times higher and 26 times higher than summer and winter background levels, respectively, and 90 times higher than summer surface waters. Such enrichment is higher than values recorded previously for the Bonney

131 Coast. Overall, average shelf values during months of strong upwelling were 6.1 times higher in NOx and 4.6 times higher in phosphate than during winter months. Determining nutrient enrichment during upwelling events compared to background levels allows for better understanding of the impact of such upwelling events on shelf ecosystems and primary productivity.

Upwelled water was, however, relatively low in silicate and had very low Si* values, which are both a signature of Southern Ocean water masses. Hydrographic and nutrient properties of upwelled water, as well as the initial depth of such water on the slope, indicates that South Australian Basin Central Water, transported west along the slope by the Flinders Current, is brought onto the shelf during strong upwelling events. This is the first study to match the specific hydrographic and nutrient properties of upwelled water to water transported by the Flinders Current. Furthermore, understanding the source water masses of upwelled water will provide additional information on nutrient characteristics being supplied to the shelf. South Australian Basin Central Water forms at the Subtropical Front south of Australia, and this region’s major, minor and trace nutrient characteristics have been more extensively studied than those of the Kangaroo Island upwelling region. Assuming such nutrient characteristics to future upwelling events on the Lincoln Shelf may be very useful in predicting ecosystem responses when upwelling occurs.

Acknowledgements. This research was supported by data collected through the Southern Australian Integrated Marine Observing System (SAIMOS). We thank the captain and crew of the R. V. Ngerin during numerous research voyages through all kinds of weather. Stable isotope analyses were done with the assistance of Kerry Klassen and April Vuletich in the Queen’s Facility for Isotope Research. Support of these analyses was from grants from the Natural Science and Engineering Research Discovery program, the Canadian Foundation for Innovation and the Ontario Innovation Foundation.

132 4.7. References

Bibby, T. S. and Moore, C. M. (2011), Si 1 icate:niträte ratios of upwelled waters control the phytoplankton community sustained by mesoscale eddies in sub- tropical North Atlantic and Pacific, Biogeosciences, 8, 657-666.

Bye, J. A. T. (1972), Oceanic circulation South of Australia, in Antarctic Oceanology II: The Australian-New Zealand Sector, Antarctic Research Series, vol. 19, edited by D. E. Hayes, pp. 95-100, AGU, Washington D. C.

Bye, J.A.T. (1983), Physical oceanography. In: Tyler, M.J., Twidale, C.R., Ling, J.K. and Holmes, J.W. (Eds.), Natural History o f the South East, Royal Society of South Australia.

Cirano, M. and J. F. Middleton (2004), Aspects of the mean wintertime circulation along Australia’s southern shelves: Numerical studies, Journal of Physical Oceanography, 34(3), 668-684.

CSIRO (2012), Commonwealth Scientific and Industrial Research Organisation. Data provided by CSIRO and reproduced with the permission of CSIRO, accessed through the Data Centre: http://www.marine.csiro.au/datacentre/

Franck, V. M., Brzezinski, M. A., Coale, K. H. and Nelson, D. M. (2000), Iron and silicic acid concentrations regulate Si uptake north and south of the Polar Frontal Zone in the Pacific Sector of the Southern Ocean. Deep-Sea Res. II47, 3315-3338.

Gill, P. C., Morrice, M. G., Page, B., Pirzl, R., Levings, A. H. and Coyne, M. (2011), Blue whale habitat selection and within-season distribution in a regional upwelling system off southern Australia, Marine Ecology Progress Series, 421, 243-263.

Gruber, N. (2008), The marine nitrogen cycle: Overview of distributions and processes. In: Nitrogen in the Marine Environment, edited by D. G. Capone, D. A. Bronk, M. R. Mulholland and E. J. Carpenter, Elsevier, Amsterdam, 1-50.

133 Hahn, S. D. (1986), Physical structure of the waters of the Southern Australian continental shelf, Technical Report 45, Flinders Institute for Atmosphere and Marine Sciences, Flinders University of South Australia, Australia, 284pp.

Hassler, C. S., M. Sinoir, L. A. Clementson and E. C. V. Butler (2012), Exploring the link between micronutrients and phytoplankton in the Southern Ocean during the 2007 Austral Summer, Frontiers in Microbiology, 3, Article 202, 26pp.

HATCH (2003), HATCH Company/Lachat Instruments Quikem Methods, Lachat Instruments, Colorado, USA.

IMOS (2012), OceanCurrent - Ocean Surface Currents and Temperature, http://oceancurrent.imos.org.au, accessed 15/11/12.

Kämpf, J., M. Doubell, D. Griffin, R. L. Matthews and T. M. Ward (2004), Evidence of a large seasonal coastal upwelling system along the southern shelf of Australia, Geophysical Research Letters, 31(9), article L09310.

Kämpf, J. (2010), On preconditioning of coastal upwelling in the eastern Great Australian Bight, Journal o f Geophysical Research, 115, Cl 2071, 11pp.

Lai, X., K. Norisuye, M. Mikata, T. Minami, A. R. Bowie and Y. Sohrin (2008), Spatial and temporal distribution of Fe, Ni, Cu and Pb along 140 degrees E in the Southern Ocean during austral summer 2001/02, Marine Chemistry, 111(3- 4), 171-183.

Lewis, R. K. (1981), Seasonal upwelling along the southeastern coastline of South- Australia, Australian Journal of Marine and Freshwater Research, 32(6), 843- 854.

Li, J. and A. J. Clarke (2004), Coastline direction, interannual flow and the strong El Nino currents along Australia’s nearly zonal southern coast, Journal of Physical Oceanography, 34, 2373-2381.

Longhurst, A. (1998), Ecological Geography o f the Sea, 398 pp., Academic Press, San Diego.

134 Lourey, M. J., J. R. Dunn and J. Waring (2006), A mixed-layer nutrient climatology of Leeuwin Current and Western Australian shelf waters: Seasonal nutrient dynamics and biomass, Journal of Marine Systems, 59, 25-51.

McCartney, M. S. and K. A. Donohue (2007), A deep cyclonic gyre in the Australian-Antarctic Basin, Progress in Oceanography, 75(4), 675-750.

McClatchie, S., J. F. Middleton and T. M. Ward (2006), Water mass analysis and alongshore variation in upwelling intensity in the eastern Great Australian Bight, Journal of Geophysical Research-Oceans, 11 l(C 8), article C08007.

McLeay, L. J., S. J. Sorokin, R J. Rogers and T. M. Ward (2003), Benthic Protection Zone of the Great Australian Bight Marine Park: 1. Literature Review, 70 pp., South Australian Research and Development Institute (Aquatic Sciences), West Beach, South Australia.

Middleton, J. F. and M. Cirano (2002), A northern boundary current along Australia’s southern shelves: The Flinders Current, Journal o f Geophysical Research- Oceans, /07(C9), article 3129.

Middleton, J. F. and G. Platov (2003), The mean summertime circulation along Australia’s southern shelves: A numerical study, Journal o f Physical Oceanography, 33(11), 2270-2287.

Middleton, J. F. and J. A. T. Bye (2007), A review of the shelf-slope circulation along Australia’s southern shelves: Cape Leeuwin to Portland, Progress in Oceanography, 75(1), 1-41.

Middleton, J. F., C. Arthur, P. Van Ruth, T. M. Ward, J. L. McClean, M. E. Maltrud, P. Gill, A. Levings and S. Middleton (2007), El Nino effects and upwelling off South Australia, Journal o f Physical Oceanography, 37(10), 2458-2477.

Petrou, K., C. S. Hassler, M. A. Doblin, K. Shelly, V. Schoemann, R. van den Enden, S. Wright and P. J. Ralph (2011), Iron-limitation and high light stress on phytoplankton populations from the Australian Sub-Antarctic Zone (SAZ), Deep-Sea Research II, 58, 2200-2211.

135 Richardson, L. E., T. K. Kyser, N. P. James and Y. Bone (2009), Analysis of hydrographic and stable isotope data to determine water masses, circulation, and mixing in the eastern Great Australian Bight, Journal of Geophysical Research, 144, C l0016, 14pp.

Ridgway, K.. R. and S. A. Condie (2004), The 5500-km-long boundary flow off western and southern Australia, Journal o f Geophysical Research-Oceans, 709(C4), article C04017.

Rochford, D. J. (1986), Seasonal-changes in the distribution of Leeuwin Current water off southern Australia, Australian Journal o f Marine and Freshwater Research, 37( 1), 1-10.

SAIMOS (2012), Cruise data, Southern Australian Integrated Marine Observing System, SARDI Aquatic Sciences, West Beach, South Australia, 5022, Australia.

Sarmiento, J. L., N. Gruber, M. Brzezinksi and J. Dunne (2004), High latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 426, 56-60.

Schahinger, R. B. (1987), Structure of coastal upwelling events observed off the southeast coast of South-Australia during February 1983-April 1984, Australian Journal of Marine and Freshwater Research, 3<3(4), 439-459.

Sohrin, T., S. Iwamoto, M. Matsui, H. Obata, E. Nakayama, K. Suzuki, N. Handa and M. Ishii (2000), The distribution of Fe in the Australian sector of the Southern Ocean, Deep-Sea Research I, 47, 55-84.

Van Ruth, P. D., G. G. Ganf and T. M. Ward (2010a), Hot-spots of primary productivity: An Alternative interpretation to Conventional upwelling models, Estuarine, Coastal and Shelf Science, 90, 142-158.

Van Ruth, P. D., G. G. Ganf and T. M. Ward (2010b), The influence of mixing on primary productivity: A unique application of classical critical depth theory, Progress in Oceanography 85, 224-235.

136 Ward, T. M., L. J. McLeay, W. F. Dimmlich, P. J. Rogers, S. A. M. McClatchie, R. Matthews, J. Kämpf and P. D. van Ruth (2006), Pelagic ecology of a northern boundary current system: effects of upwelling on the production and distribution of sardine (Sardinops sagax), anchovy (Engraulis australis) and southern bluefin tuna (Thunnus maccoyii) in the Great Australian Bight, Fisheries Oceanography, 15(3), 191-207.

137 Chapter 5.

Research Conclusions

The aims of this thesis are twofold -

1) To characterise the regional water masses and connectivity of ocean currents along the southern margin of Australia, and

2) Within this regional context, determine water masses and upwelling characteristics of the Kangaroo Island upwelling region, Lincoln Shelf, South Australia.

Central to these aims was the Flinders Current and subantarctic water transported by it. By first characterising the hydrographic properties and depth distribution of the Flinders Current along the southern margin, it is then possible to identify the influence of subantarctic water on the Lincoln Shelf during periods of upwelling. Traditional hydrographic analyses of temperature, salinity and dissolved oxygen are combined with oxygen and hydrogen stable isotopes of seawater, and major nutrients nitrate, phosphate and silicate, to determine water mass characteristics on both a regional and local scale.

This thesis is composed of three linked studies. The first identifies water masses and regional connectivity along the southern margin of Australia; the second determines water masses on the Lincoln Shelf, South Australia during all times of the year; and the third focuses specifically on upwelling events, utilising the knowledge of regional water masses from the first study and shelf water masses from the second study to understand how the hydrographic and nutrient characteristics of upwelled water compare to background conditions on the Lincoln Shelf. The key findings and conclusions of each study are discussed in the following sections.

138 5.1 Key findings

5.1.1 Regional water masses and Flinders Current connectivity

This study is the first to identify water masses along the southern Australia continental shelf and slope. The depth distribution of water masses and knowledge of water mass source regions has made it possible to determine which water masses are transported by the Flinders Current, constraining the hydrographic properties and depth range of the current.

Four water masses are identified in the top 1000 meters water depth (mwd) along the southern Australian margin. Three water masses correspond to those identified in previous water mass studies in the region, Subtropical Surface Water (STSW), Tasmanian Subantarctic Mode Water (TSAMW) and Tasmanian Intermediate Water (TIW), and one water mass is newly identified and named, South Australian Basin Central Water (SABCW).

STSW is transported east by the Leeuwin Current system (Figure 5.1) and is modified by heating and evaporation along the subtropical continental shelf; SABCW is formed at the surface at the Subtropical Front (~40-45°S) south of Australia; TSAMW is formed at the surface within the Subantarctic Zone southwest of Tasmania; and TIW is formed west of Tasmania when cold, fresh AAIW from south of the Subantarctic Front mixes with older, more saline AAIW from the Tasman Sea. The Southern Ocean source regions of SABCW, TSAMW and TIW identify that the Flinders Current transports these water masses north and then west along the southern Australian margin.

The interface between subtropical water transported by the Leeuwin Current system and subantarctic water transported by the Flinders Current is between STSW and SABCW at ~250 mwd, respectively labelled A and B in Figure 5.2. The depth of this interface varies seasonally and can be as shallow as 150 mwd in the Kangaroo Island upwelling region and off western Tasmania. The presence of Flinders Current water at these depths is much shallower than previously thought (Middleton and Bye, 2007).

139 30°S - West Leeuwin AUSTRALIA Current South Australian Australian _ _ Current Current TW ) ----- \ r ^ S T S W SICW

f/SICW Zeehan South Indian t Flinders Current Current Ocean Current SABCW, TSAMW, TIW ICW * subduction? j SABCW formation Subtropical Front Tasman Outflow saline AAIW TIW / Subantarctic Zone formation TSAMW^ Antarctic Circumpolar Current formation fresh AAIW

110°E 130°E 140°E 150°E Longitude (°E)

Figure 5.1. Schematic of currents and water masses south of Australia. Water masses: TW - Tropical Water, SICW - South Indian Central Water, ICW - Indian Central Water, STSW - Subtropical Surface Water, SABCW - South Australian Basin Central Water, TSAMW - Tasmanian Subantarctic Mode Water, TIW - Tasmanian Intermediate Water, AAIW - Antarctic Intermediate Water.

Summer TEMPERATURE

110°E 120°E 130°E HO°E 150°E SALINITY

Figure 5.2. Averaged summer (November to April) temperature (°C) (a) and salinity (b) along the southern margin from 110°E to 147°E. A-D are water masses: A - STSW, B - SABCW, C - TSAMW, D - TIW, D! - low salinity core of TIW. STSW is transported east by the Leeuwin Current System while SABCW, TSAMW and TIW are transported west by the Flinders Current.

140 The relative abundance of silicate to nitrate, Si*, is used as a proxy to help differentiate the Leeuwin Current and Flinders Current. Surface Si* values in the Southern Ocean are the lowest in the global ocean (Sarmiento et al., 2004), therefore very low Si* values (<-5 pmol/L) are used as a signature for the Flinders Current. In contrast, tropical water on the western margin of Australia has relatively high surface water silicate values (Thompson et al., 2011), therefore high Si* values (>0 pmol/L) are used as a signature of the Leeuwin Current system. Si* values support the above water mass distribution and interface depth between the two current systems.

Stable isotopes of seawater decrease with increasing depth, with STSW having highest isotope values and TIW having lowest isotope values based on the latitude of their formation at the sea surface. A linear salinity-isotope relationship between STSW, SABCW and TSAMW is a result of their similar formation region and mixing patterns south of Australia. In contrast, the two AAIW types that mix to create TIW were formed south of 60°S in regions remote to southern Australia, and a change in the salinity-isotope relationship is evident between TSAMW and TIW as a result.

This study is also the first to confirm that the Flinders Current continues north along the western margin of Australia as the Leeuwin Undercurrent. Stable isotope values at the STF within the South Australian Basin and west of Perth match well with isotope values along the southern margin, supporting the movement of water masses from the Subantarctic Zone, west along the southern margin by the Flinders Current and north along the western Australian margin by the Leeuwin Undercurrent (Figure 5.1). In contrast, differences are seen in isotope values within the Tasman Sea and at North West Cape, Western Australia, due to different water mass influences external to southern Australia.

5.1.2 Lincoln Shelf water masses

This study is the first to identify water masses for the whole of the Lincoln Shelf, recognize the origin of upwelled waters, and better define mixing and circulation between water masses. The use of stable isotopes of oxygen and hydrogen together with hydrographic data has made it possible to identify source regions of water masses and the formation of mixed waters, and better define the spatial distribution of water masses on the shelf.

141 Five new water masses are defined on the shelf (Figure 5.3). Three water masses with specific source regions and formation modes are present year round, and two water masses form by variable mixing on the shelf during summer. The three source water masses are:

• Slope Water (SW) - a low temperature, salinity and stable isotope end member that is found year round at depths greater than 180 mwd, and is sporadically upwelled onto the shelf during summer from depths as great as 300 mwd. During such strong upwelling events SW is sourced from SABCW, a water mass formed at the Subtropical Front south of Australia that is transported west along the slope by the Flinders Current. It is upwelled either through submarine canyons to the south of Kangaroo Island or is upwelled along the Bonney Coast and transported onto the Lincoln Shelf by a westward flowing shelf current.

• Subtropical Surface Water (STSW) - a shelf water mass with intermediate salinity and stable isotope values that is transported year round into the region by the eastward flowing South Australian Current. It is well mixed on the shelf down to depths of 180 mwd during winter, and is displaced at depth by the inflow of SW during summer upwelling events.

• Evaporated Water (EW) - a high salinity and stable isotope end member that is formed locally at the surface on the inner shelf and within Spencer Gulf from high evaporation and heating during summer. It outflows from Spencer Gulf onto the shelf at depth during winter.

The use of stable isotopes has made it possible to identify two additional mixed water masses, which are produced on the shelf during summer. The first is Mixed Slope Water (MSW). Despite having moderate temperature and salinity, stable isotope values indicate it is formed by mixing of S W with STSW on the shelf during upwelling events, as it lies on the isotopic mixing line between these two source water masses. Spatial distribution of this water mass in relation to SW suggests that SW is transported west towards Eyre Peninsula and north into the mouth of Spencer Gulf, and vertical mixing allows nutrients of SW to be brought into the photic zone to be utilised by primary producers.

142 The second mixed water mass is Cooled Evaporated Water (CEW), a modification of EW that is identified from stable isotope values identical to EW, but with lower temperatures and salinities. It is found near the seafloor during summer, predominantly around the western tip of Kangaroo Island, however the identical stable isotope signature to EW confirms that this water mass had recently been evaporated at the surface. Early stages of vertical mixing between surface EW and cooler, fresher bottom waters result in equilibration of temperature and salinity but preservation of the more conservative stable isotope signatures, to form CEW. This is the first recognition of vertical mixing through a stable stratified water column in the region, and could be important for vertical mixing of nutrients both on the shelf and in Spencer Gulf. A possible mechanism to achieve this phenomenon is double diffusion by salt fingering, whereby hot, salty surface water (EW) can mix with near bottom cool, fresh water. Double diffusion may be a mechanism to bring nutrients from SW and MSW into the photic zone (M. Doubell, pers. comm., July 4, 2013).

, South Australia

>pencer< G u lL j

Longitude Eyre Peninsula Kangaroo Island

Evaporated Water patches

vertical mixing Subtropical Surface Water double diffusion

Cooled Evaporated Water westward flow, mixing

Mixed Slope Water Slope Water Kangaroo Island Upwelling I I 135.5°E Longitudinal Section 136.5°E

Figure 5.3. Schematic of summer shelf water masses during periods of upwelling. Slope Water mixes with Subtropical Surface Water to form Mixed Slope Water as it flows west and mixes vertically. Patches of Evaporated Water form at the surface, and possible double diffusion processes mix it with bottom waters to form Cooled Evaporated Water. Top insert shows location of cross section A-Al on the Lincoln Shelf; GSV - Gulf St. Vincent.

143 5.1.3 Upwelling characteristics of the Kangaroo Island upwelling region

This study is the first to define the characteristics and source depth of upwelled water in the Kangaroo Island upwelling region, by analysing four strong upwelling events and several weak upwelling events between February 2008 and February 2011. It is also the first study to record significant nutrient enrichment during upwelling events, and then compare such enrichment to conditions during non- upwelling periods.

The upwelling signature on the Lincoln Shelf is less than 13.5°C and 35.3 during weak upwelling events and less than 12.5°C and 35.2 during strong upwelling events. Strong upwelling events occurred during February and March 2008 and February and March 2010. The coldest and freshest water recorded on the shelf during the study period had temperatures/salinities of 10.4°C/34.85 (Figure 5.4a), sampled at 100-110 mwd during February 2008 and March 2010. Temperature, salinity and nutrient data from this study suggest that at least 300 mwd is needed to supply such cold, fresh and nutrient-rich water to the shelf, indicating water is upwelled over depths of 200 m or more. The average depth of this water recorded in slope profiles from historic cruise data is 380 mwd, therefore isotherms can be raised significantly during upwelling events.

Entry of upwelled water onto the shelf remains uncertain during some upwelling months due to the spatial distribution of data, but appeared to be coming from the Bonney Coast during February 2008 and from the shelf edge south of Kangaroo Island during February 2010. SST images during March 2010 suggest that upwelled water from the Bonney Coast spreads west along the shelf edge south of Kangaroo Island, and could have sourced upwelled water on the Lincoln Shelf during this time. If this is the case, the three upwelling centres along the South Australian coast - the Bonney Coast, Kangaroo Island, and western Eyre Peninsula - could all be sourced from the Bonney Coast.

Upwelled water supplies significant macro-nutrients to an otherwise oligotrophic shelf. During strong upwelling events water upwelled onto the shelf had the highest values of NOx, phosphate and silicate. The most nutrient-rich water on the shelf in March 2010 had NOx and phosphate concentrations of 13.35 and 0.94 pmol/L, respectively (Figure 5.4b). These values were 18 times higher and 26 times higher

144 than summer and winter background levels, respectively, and 90 times higher than summer surface waters. Such enrichment is higher than values recorded previously for the Bonney Coast. Overall, average shelf values during months of strong upwelling were 6.1 times higher in NOx and 4.6 times higher in phosphate than during winter months. Determining nutrient enrichment during upwelling events compared to background levels allows for better understanding of the impact of such upwelling events on shelf ecosystems and primary productivity.

(a) T-S at 100-120m (b) NOxvs phosphate

winter

o weak upwelling □ strong upwelling

0.4 0.6 Salinity Phosphate (pmol/L)

Figure 5.4. Properties of water on the Lincoln Shelf, during winter months, summer months with no upwelling, summer months with weak upwelling and summer months with strong upwelling, comparing a) temperature vs salinity (T-S) of water at 100-120 m water depth, and b) NOx vs phosphate (all depths).

Upwelled water was, however, relatively low in silicate and had very low Si* values, which are both a signature of Southern Ocean water masses. Hydrographic and nutrient properties of upwelled water, as well as the initial depth of such water on the slope, indicates that SABCW, transported west along the slope by the Flinders Current, is brought onto the shelf during strong upwelling events. This is the first study to confirm that Flinders Current water is upwelled onto the shelf during strong upwelling events.

145 5.2 Research Implications

5.2.1 Identification of the Flinders Current

While previous studies have focused on water mass formation in the Indian Ocean and within the Subantarctic Zone south of Australia, none have focused on the shelf and slope of the southern Australian continental margin. Hydrography and chemistry of the water in this extensive region is important for slope habitats as well as shelf ecosystems in upwelling areas such as Kangaroo Island and the Bonney Coast.

Based on the characteristics and distribution of water masses, their formation region, and the regional ocean circulation, SABCW, TSAMW and TIW all form south of Australia and are transported north towards the Australian coastline and west along the slope by the Flinders Current (Figure 5.1). It is important to determine both the depth range and water masses of the Flinders Current to understand the influence of Southern Ocean water masses on shelf and slope ecosystems along the southern margin. The Flinders Current transports SABCW that is found as shallow as 150 mwd on the western Tasmanian slope and off Kangaroo Is. during summer (Figure 5.2) - much shallower than previously thought. The presence of the Flinders Current at such shallow depths means that deep upwelling off Kangaroo Island and the Bonney Coast transports SABCW of high nitrate, phosphate and relatively low silicate onto the shelf.

Identifying the influence of Tasman Outflow water on TIW properties along the western Tasmanian slope confirms a link between the Tasman Sea and southern Australian margin. Warm, saline Antarctic Intermediate Water from the Tasman Sea is incorporated into TIW and then transported west along the southern margin by the Flinders Current and north up the western Australian margin by the Leeuwin Undercurrent. Such water mass movement implies that water from the Pacific Ocean is transported into the Indian Ocean at intermediate water depths, providing a pathway of intermediate water between ocean basins. This link has implications for water mass movement on a global scale.

5.2.2 Use of stable isotopes in shelf water mass studies

The use of stable isotopes as conservative tracers in water mass analysis provides additional information that temperature and salinity alone cannot supply. Stable 146 isotopes are able to identify the source region of a water mass, even after mixing has begun, and can recognise the formation of mixed waters. For example, surface waters on the Lincoln Shelf, identified as STSW, have the same isotope signature and hence the same source region during both summer and winter. The South Australian Current transports STSW from the Great Australian Bight to the Lincoln Shelf, therefore the presence of STSW year round in the region confirms that the current is flowing in summer as well as winter, which has not been previously identified. Despite previously recognised high seasonality in the region, STSW and EW were found on the Lincoln Shelf during summer and winter, and while SW was not present on the shelf during winter, stable isotopes identified that the water mass was present year-round at depths greater than 180 mwd on the slope.

The input of SW onto the Lincoln Shelf during summer resulted in more complex water mass interactions. Isotopes were able to identify two mixed water masses, MSW and CEW (Figure 5.2), which were not distinguished by temperature and salinity. These mixed water masses provide a great deal of information about mixing of water and nutrients on the shelf. They are also able to identify the path of nutrient- rich water away from the upwelling centre. This study further confirms that during mixing of different water masses, isotopes can retain their signature longer than temperature and salinity, making isotopes invaluable tracers in shelf areas with complex oceanography and mixing.

5.2.3 Identifying upwelling characteristics

Defining the characteristics of seasonal upwelling events on the Lincoln Shelf is valuable for predicting shelf ecosystem responses and increases in primary and secondary productivity, for both ecological management and commercial fisheries productions. Major nutrient concentrations recorded during upwelling events are significantly higher than concentrations measured during other times of the year (Figure 5.4b). Nitrate concentrations up to 90 times higher in upwelled water compared to surface waters means that upwelling supplies significant nutrients to shelf waters, which are oligotrophic and devoid of nutrients during most of the year. Knowledge of nutrient ratios is also important to determine ecosystem community structure. High nitrate, low silicate upwelled water can result in silicate being a limiting nutrient, which has implications for diatom production on the shelf.

147 Defining the source water mass of upwelled water using hydrographic properties and the depth of upwelling can provide additional information on nutrient characteristics being supplied to the shelf. South Australian Basin Central Water forms at the Subtropical Front south of Australia, and this region’s major, minor and trace nutrient characteristics have been more extensively studied than those of the Kangaroo Island upwelling region. Knowledge of minor and trace nutrients in Southern Ocean waters can be extrapolated to the Lincoln Shelf where no knowledge exists of minor and trace nutrient enrichment during upwelling events. Assuming such nutrient characteristics to future upwelling events on the Lincoln Shelf may be very useful in predicting ecosystem responses when upwelling occurs.

5.3 Future research

Recommendations for future research include:

1. More detailed analysis of the area south and west of Tasmania will help constrain the Flinders Current pathway from the Subantarctic Zone to the southern Australian margin. Minimal observational and stable isotope data for the Subantarctic Zone also restricted understanding of the formation region of SABCW and how it is incorporated into the Flinders Current.

2. Additional stable isotope analyses for regions external to the southern margin would help confirm continuity of the Flinders Current between the Southern Ocean and western Australian margin. In the current study data coverage was limited to two stable isotope profiles in the Subantarctic Zone and two profiles in the Tasman Sea. Additional isotope analyses would also help confirm isotope signatures of TSAMW and TIW, which could then be better compared to Subantarctic Mode Water and Antarctic Intermediate Water in other regions. Stable isotope data coverage close to the Subantarctic Front could help clarify the influence of polar water on TSAMW and TIW formation.

3. Minor and trace element data for the Lincoln Shelf during both upwelling and non-upwelling periods would broaden current understanding of upwelling characteristics and upwelling influence on shelf ecosystems. In addition, carbon and nitrogen stable isotope data would link the

148 oceanography with biological systems, providing a more robust understanding of ecosystem responses to upwelling.

5.4 Concluding statement

Determining the hydrography and depth distribution of the Flinders Current is imperative to understanding the influence of Southern Ocean water on shelf ecosystems in the Kangaroo Island upwelling region, South Australia. The shallowest water mass of the Flinders Current, South Australian Basin Central Water, is present at -250 mwd along the southern margin, and upwells onto the Lincoln Shelf during strong summer upwelling events. Southern Ocean water with high nitrate and phosphate, but relatively low silicate, is supplied to a shelf that is devoid of nutrients during most of the year. Defining the chemistry of upwelled water makes it possible to understand biological responses to upwelling events, providing a link between the physical and biological systems.

149 5.5 References

Middleton, J. F. and J. A. T. Bye (2007), A review of the shelf-slope circulation along Australia’s southern shelves: Cape Leeuwin to Portland, Progress in Oceanography, 75(1), 1-41.

Sarmiento, J. L., N. Gruber, M. Brzezinksi and J. Dunne (2004), High latitude controls of thermocline nutrients and low latitude biological productivity, Nature, 426, 56-60.

Thompson, P. A., K. Wild-Allen, M. Lourey, C. Rousseaux, A. M. Waite, M. Feng and L. E. Beckley (2011), Nutrients in an oligotrophic boundary current: Evidence of a new role for the Leeuwin Current, Progress in Oceanography, 91, 345-359.

150 Appendix A - Data file for the southern margin (used in Chapter 2) 3 T 3 ge 2 •4 of O — tO c s . E z o ~ O £ II - 5 LO o -O J3 E ill ^ c c2 re >» j at oj ii sS 3 J TO £ QJ 3 E £ 6 OJ rö O op är > 3 3 3

s d d d l d d 00 d l d rn d d o rsi o rH HrH rH \O CTv ON Q\ cn * * 00 fN d l r< £3 LD O LO LO fNfNfNfNfNjrMfNfNfNJfNfNJfNfNrNfN fN|fNfNfNfNjfNfM(NfNfNfN|fN(NfNlfNfNfNfMfNJfNfN(NrM(NfM o o o o o oo o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o - r r r rI — r r r r r r r r r r r r r r r r r r rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH r—I rH r-I rH rH rH r-I H H H H H H H H H H H H H H H H H H H H H H H H rH rH rH rH rH rH rH rH rH rH rH d ö rH o o o rH o o ö rHLDLDLDLDLDLDLDLDLDLDLDLDfNfNfNfNfNfNfNfNfNfNfNfNLD o o rH o ö o rH ö o ö rH o d rH o ö D c L rH ö D o L o rH o D n L r rH n iD r D n rH r L n D r rH L n D r rH o L r D n L r rH D n r L n D f L o D r i L Y D f L n r iD m m m m m m m rnrnrnrnrorornrnrornrnrnrvr^r^r^r^r^r^r^r^r^r^r^r^ m m m m m OLL /L OL OLOLOCOLOLOt-OLOLOLOLOl-Ot/)LOLOLOLOLO LO LO LO t/)LO LnLO LO cncOLOLOLOLntOLOLOLOLOLOLntOLOLOLOLOCOLOLOLOcOLOCO fMfSfNfN(NfMfNfNfNOJ(NfNfNrvJ(NfNfNrNlfNJfNfNl(NfNfMfM o o o o o o o o o o o o o o o o o o o o o o o o o o o o or o H or o o H or o o H or o H o or o H or Hr r o H r o H H r H o rr H r o H r H o H r H o rr H r o H r H H o r H rr H r H H r H r Hr r H r H H r H r Hr r H r H H r H r Hr r H Sr J Hr Hr HCDCDCDCDCDCDCDCDCDCDCDCDfNrNfNfNfNfNrNfNfNfNfNfNm r Hr Hr Hr Hr Hr Hr Hr Hr Hf N O O O O O O O O O O O O O O O O O O O O O O O O O r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H r H >>>>>>>>>>>>>>>>>>>>>>>>> I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I O O fN d l LO 00 CD N fN fN LDLO m m m £5 s 00 m 00 d l d l S d d 00 fN 00 s o o o 00 00 fN d l rn fN fN t a fN d l a> rH o d a> d o d l fN m fN d l g d 0OS O 00 00 fN LD d 00 o rH fN g d l fN cn fN dd LD LD rsi 00 LD fN LD d o a> fN LO fN LD 00 m m m m m m m m LD LD LD rn CD CD rsi LD 00 CN LD O 2 rsi 00 fN m n r LD 00 fN LD £ r-H DLD LD £ rn 9 9 3 £ CN fN rn LD CD CD d O CN d O HrH rH LD Si fN LD fN LD 00 CD in fN fN fN m NfN fN 123 LD rn CD CD fN g in LD LD fN O LD fN LD rn fN m 00 LD 00 (N 147 • 3S° 5 » > ■ n s i rn fN q fN LD q q q n N OCD LO fN rn D D LD CD ID LD CD IN q m H H rH rH rH H N fN fN rH rN H rH rH N N fN fN fN D 0 CD 00 CDLD LD LD N N fN fN LDfN LD LD D D 00 LD LD ID fN 00 o o o N N N . H ^ ^ ^ CD O rN ^ rH ^ 297 ss in o m D n r ID d l 0 l< 00 oo 497 ^ fN 747 s s m LD m rsi o rH d 5 n r 2 fN IN LD m d rH 2 m LD 997 d Si dddddddd o s d OfN VO d n r d m 00 rn rn fN LD CD o m m m m m m m m m m LD a> DLD ID iSi Si rH 00 fN LD LD O I dd rn d fN rH ddddddddd m CD g fN LD LD LD LD JN RJ d LD LD d LD NfN fN CD d n r m rn rH s JN fN LD S 00 D CD ID Ng JN d Si d

CD CD CD 00 CD CD CD CD 00 CD LO 00 QJ LO s § £ s fN CD CD m m 8 m m CD 00 £ 8 2 4 d d d d d d HH H H r-i d d d d d d d d r-i r-H r -i rsi

00 CD r n LO CD CD i n rH fN 00 CD cd fN m 8 CD CD 00 £ g 2 s o a t-H rH s s fN 00 T-i r -i CD d r n CD 00 CD fN rsi rsi rsi CD 00 CD 4 rH rH rH rH rH rH rH fN fN tH fN rsi m fN fN CD I - g CD o r n r n i n fN r n LO CD fN CD r** fN 00 CD r n m 00 00 m 4 o ü h d rH rH rH T—l rH rH d d tH rH rH rH rH tH rH tH d r n r n rsi rsi fN

O ~ LD CD o fN 00 Ln 00 rH o m m CD o o o 2 fN fN O o o o m m a 8 2 2 2 ä ? 8 a LO d d o d o o o o o o o d d d d O o o o o o o o d d O O d

00 CD fN 00 m CD CD CD CD i n m m 00 m tH 00 m r - o w E 8 m LO CD So 00 00 CD m m m m s s 00 8 fN 8 m 8 m CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD r^‘ LO uS CD CD CD 11 fN fN fN fN fN (N fN fN fN fN fN fN fN fN fN fN rN fN fN rsi fN fN fN fN fN rsi fN fN

c d rs j m m t—< £ i CD ^ CD CD i n II Ln in CD rn rn o 3 CN fN CN fN fN

>■ f^ooOT-HcDr^r^cDcDCDoommmmuncDmcDrMr^cDLD^-cDOcDro CDCDCDOCDr^CDLOLD^mOOOOOOOCXDCDLn^^cDCD^-mm 75 co mmmrommrnmmmmmmmmmmmmrnmmmmrommm

cü o» cc in rH CDmoomfNfN^^^^^. fN rH rH rH 0)es TO E £ ~ S i. S^°2g°gooS|sS£32gg|g|§g|co«§gg 0

CD CD CD CD CD rn rn rn rn ro m m m m m 1 l l 00 00 00 00 00 ^ Q m cn m cn m oooooooooooooooooooooooooooo TO rHrHTHrHtHrHTHrHTHrHTHTHrHrHTHTHrHTHTHrHTHTHTHrHrHrHrHrH QJ oooooooooooooooooooooooooooo fNMnjfNfNfM(N(NfNI(N(NfNfNfNrMfNfNfN(NfNfNlfMfNfNfNfNfM(N

rcDcDcDcDcDcDcDcDcDcDcDmmmromrommmrommr^r^r^r^r^ ^ r ^ r ^ r ^ r ^ r ^ r ^ r ^ r ^ r ^ - r ^ c n c n cn cn cn cn cn cn cn cn cn cn ^ ^ ÖÖÖÖÖÖdÖÖÖÖfHrHrHrHfHrHrHrHrHrHrHrHÖÖÖÖÖ C O rHrHrHrHrHrHrHrHrHrHrHrHrHrHrHrHtHrHrHrHTHrHrHrHTHrHrHTH

“O CDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCDCD B rornrnrOfY^fY^(Y^(rirnrnrncDcDcDcDcDCDCDcDcDCDCDCDooooo IDCOldcÖcÖCDCDlDcdcDldlDlDCDCDCDcdldcdtdldcÖldcriLriLricricri £TO I I I I I I I I I I I I I I I I I I I I I I I I I I I I

fNrMfNrMfNfNfNfNMfMfN^^^^^^^^^^^^OOOOMOOOO

fNfNfNfNfNfNfNfNfNrNfMfNrNfMfNNfMfNfNfNrNfNfNHHHHH oooooooooooooooooooooooooooo är >, >1 >1 >, >l >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 > oooooooooooooooooooooooooooo rH rH tH tH tH rH rH tH rH rH rH rH H rH rH rH tH rH rH rH rH r—I rH rH rH rH rH rH oooooooooooooooooooooooooooo 5 fN(NfNfNfN(NfNfVIOJNfNfNfNfNr\|fM(NfNfMfSfMfNfNfNrjfNfN|fM LOLOl/nLOLOLOLOLOLOLOLOl/ncOCOLOCOC/nc/ncOC/nLOCOLOLOLOLOLOLO •i

-I a 2 1

O„5 o Z E 3

x T o oo m ^ m rsi oo ^ ’I ’I rny)Oroi\m^yj^ oomLnoor^Lno^ To £ rN h ö m rn fN rN d 9 9 corNrnrsiT-H

o 'S- s g? LO -—• 0990000900000000900000099999

> r ^ r 5 r^ooominLnrMr^moT-HrHmcoTHcrjovDO^OrHOcMmomrN^cDr^mLnm^-r^Ot-HrofNiorsjunLncDi^oorofNO^rr^oqOTHrM •) E ibiD^i/iinio^ioibr^NuiiiiibibiDiD^iiirniriidiDiDiDisisrs IS (NfMNfNfNfNjfNIMfNfMfNfMfNfNlNrMfNfNfMfNIfNfNfNfNfNfNfNfN

in^rO^^fNOO^'tHN^LnfNOlOlOlDHmfNlDtM^^^fNCO ininrv'tmcorvjincorn^iniocoNrnLno^^^miDfNdiro^H^ £ i ^r^fsifOfNiD^fNinaiNHkdi/iinH^iniri^oöfNfNi/iri-fNii/iH ^■^-^■mm^-^-^frHcnoomLn^j-^-LOLn^-^j-THLncocD^-mLnr-HO oii 3 (NfMfNfMNfNfNfMMHHfNfNfNfMrgfMfNfNIfNHHHfNfNHHH

CDOroLnf^Lnor^mrNixiOrHcDrNir^r^^ooomcDr^LncDrtLnr^ »Hcrior>;'X>rnfNo^rrorooo^r^-fNC)i^cDLnOi-HT-HrHi^LnLnLnLn

03 mmmmmmnomm mmmmmmmmmmmmmmmmrom CO

a3 ro o i_n m O cn co rsi Q. ° 1 ^ m 0 0 vd *"■. “ l. " 1 "I lti o w. ui. ^ l>* vo vrj o u rn t-i O. in E01 TO

£ - O O m in ro rsj o lT) t-h co ,n cnj m tH pm tH rN o *h 0

CDCDlDCDlDCDlOlOCOlOCO(NfMr'lfN(NfMfNJ(NfM(NfN|fMfNfN(N{\fN X ^ -—* rornrornrornrnrnrnroro^^^^^^^^^^M ; ^ ,7 ^ ^ ^ ^ E. mmmOOOOOOOOrsirsifNfNfN(NrMfNifNrsjfN(NrsifNrNjfNrsj I S oooooocovococDcococDcoLnLOLnLnLnLnLnLnooooooooo mmm(NrNrNfNrNrNrsl

OOOOOOOOOOOOOOOOOOOOOOOOOOOO tH tH tH tH tH tH tH rH tH tH tH tH tH tH tH tH tH rH tH r—I tH tH tH tH rH rH tH tH OOOOOOOOOOOOOOOOOOOOOOOOOOOO fNfNfNfNfNfNlfMfM(NfMfMfN|fNfMfVlfNrMfN(N(NfNfN|fNfMfMfNfNfN

_g [^«tHtHtHtHtHtHtHtH O') O') O'! O') O') O') O') O'! LT) LD LO LD LH LT) LT) LT) LO 1 ^^^iDcDcDcDcDcDcDcoo^oitno^aio^o^oiinini/i^ininininm ddddcdcdcDdcddcdddrncndcncncnrncnrnrnrnrncncncn C3 fN fN fN tH tH tH t-H t-H t-H t-H t-H t-H tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH o tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH

“D voocooooooooooooooooooooooooor^i^i^r^r^r^i^r^rN. 3 O O O LO LT) LO LO LO LO LD LD O O O O O O O O tH tH tH tH tH tH tH tH tH mrnmrnmmmmmmmmfY3 mmmmmmfNifNirsjrvjfMfNi(N

tH t—i tH t—I tH tHtHtHtHt—ItHtHt—ItHtH tH tH tH tH tH tHtHtHtHtHtHtHtH OOOOOOOOOOOOOOOOOOOOOOOOOOOO

>S' «2 < S3SS3§s S?5ES^?S^^||^|| rH rH rH N op saaasN

gooor~^iTH»Hoo(r)[^[2oq5;g[^g2^^[J’ p2^og[J’ ^ cr'<:r’ rNH H i-ifsim 'jinib^^^^HH ddH ririi^'iuiK

rsii^i^CTioor~(N— irNC\iÖdÖÖÖÖÖ»-i'—i<—i<—i °13

g - S R S S 3 S 3 S S S S i q B S osis3EgSgfC2S Zo-i £ d d d r r i OrniD,^ 00rncr''-lrNld d d d r n d ' - |L/1lß00O 3

x -g- ^ ' x! o ^ j; o q r o rH qLn'TO ;r^^;fr! c!'-!';a;rooqCTiLni^; r^oo t-t to- H 9 9 ff'^°iN H H ddddO H H ro - rM ' cm ‘ : ‘ r^o* - O ’-' d o

O I -s- 2 ä? LO — oqodddoqqqqqqqqodddddddddd

4 r^T-H^cxJOLnor^fN^-UDm^j-t-HoorNrNmLOLncxDfNOrN^tr^ £ rMmrHfNjro^3-r^.r^oooooocrit-HrsifNT-Ht-Hmrn^3’ Lnr^oocx)oooo r4r4u5u5*£>ix><£)i£>voixjr4f4r4»x>i£)vbi£><£)t£ji£>i£>UDv£> IS fNfNfNlfN(N|fNfNfNfNfNfM(NfN(NfNfMfNfNlNf'JfMfNfMfNfNfM p^.r^T-HcnfNmmooLncxDrHoor^^-rMr-HOrH^DCDcTi^rmcnr^ Si z:^LnLnT-Hi^LnrNcx)Ocr>o^fNcr>mr^OfNoo^r^r^T-HvDfNm ^^o^(NrnH^[srsci6L^Nrnd0kDrs(NHfNfNrnrsdr^d OSI a. ulu'rMfNrMfNfNfNl(NfN(NfMfNfNHfNfNfNfMfNfNfN|fNfN|fNf\| 3 cocoomH^iDH^rooiH^coQ^^oi^^fN^moirn^D .E LDLn'^DLDLnma>)OO^DUDLnLnrorom'.^oisDLn

1/1TO mmmrorornmmmrommmmmmmmmmmrnmmmm 34.48 215.47 34.41 202.86 26.97 27.10 -0.38 -0.35 -0.3 -0.8 25.37 29.56 1.67 1.93 15.1! 25.4! TO cn vd LD rN ^ L4 Q. ^ ^ O r f CT> rsl CD U • kD HD ^ in ^ fM £ r-H rH rH rH rH rH oS cri oo cxi |4 L/S (1) TO

£ - CTiai «-tr>im^’^i/ivor'»ooCTiJ^ ^

N N ^ ^ , ^^^'t'}<}^'^^'}^iN(N(NNr

^■"ifOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOOOOOOOO

_TO

ininc«c)oa)oooooooooqcx)oqoqoqooininininininininininininLn IM rnmdoioioioioioioioioioioioi[N(N(N(N(N(Nf\itN(\ilN(NN(N rH rH rH rH *—I rH rH rH rH rH t—I rH rH rH rH fN fN fN fN fN fN fN fN fN fN fN fN fN rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH r—) rH rH rH rH rH rH 0

_TO 1 P^r-^OOOCriO'lCTlO'lO'lO'lCriO'lO'lO'lrHrHrHrHrHrHrHrHrHrHrHrHrH £ fNfNmmmmmmmmmmrorommmmmmromrnrnmmmmI I I I I I I I I I I I I I I I I I I I I I I I I I I I

m m (NfNi(Nf\JfNfNfNf\|(NfM(NfNfN

HHfNfNnjfN|fNfNfN(NrJfNlfNOJrMfNfNfNfNfNfN(NfNfNfNfNfMfM OOOOOOOOOOOOOOOOOOOOOOOOOOOO >TO H" l/"> 3 *1

^LDPNJOtHTHOLnmrHT-tLDfNfS ‘irHrNrNrnr^qqqqror^i-n. 4 ^T-HrHTHrHrHrHrO^LnLnLn^^ ^^HHHHrslro^ui^aiS

rHr^vor>.(x>i^f^p^<3-r^(r>THmrH cxD^j-cxDf^m^trHT-HomfNmo o’ ? rHrHrHrHrOLntOOOrHrHfNmO 9 H(Yl frl^ fN f ) f~^ C ^ C ) C ^ ( ) rH rH rH rH rH rH PM fN O O O O O O t H rH rH rH rH

„ i O o 83sss$sssssss Z E m r o r n m ~ ~ a’ iHrHiH<-Hr\irg 3

J E 'g - r1 ^ 0 C 0 0 1 0 0 l D M U l r ' l ! S r! r| 9T7l fN **(N ° i o iNfMHHodddd?? r?'"?rN<-ii-ir-i,-ir-iddd9C?

O 'S- S 2 S 3 o h ^ S ! S ^ s ä? t o '—' oddoddoooooooo 9 9 °°°°o999999

rM^-^-^-ooCTivDCTiom^-'i-r^LniDfMvofMfNmLnCTir^T-ifMmoo^D (Nmmro^t-LDUDr^oooooooooocnO(NfNix>iX)U3iDix!rvoooooooocn i^dddddddddddddr^r^r^ddddddddddd fNfNfMfNfN(NfSfMf\lfNfNfNfNfN|fN(NfNfN(NfNfNf\|fNJfNfNfNtN(N

rsir>*oo^-c7»(NOcr>r^o^kOLncr)oor^ooLno^r^oooOfNr^THtHoo S < Lno'iouDO'H^crim^omTHr^i^DcxDoor^rNi^r^to^j-Ln^j-criLnfN cxioHOHfri^rvoi^NisH^rorsinddoiHinoidr^iDisN II HfN|(N(NM(NfNfNfNfNfNMfNfN(N|HHfNfMfNlfNjfNfN(NfMfNfMfN

CT>mcnmooooooro'X)cr»LnLnTHcr>^ovorMOO^or\i I mLOLnLnmoor'^r^^D^D^DLn^f^'romrNfNfNrHoootD^DvoLnLn <2 mmmmmmmmmm commcncncncocnmcncnrommmmm

cü oj ^ f>4 r*a fN pv. r o VD rH ^ Ln Ln uS rn fN r-i o ai in ^ m a n M. M. ^ N. t ri oi oo rf) \o ISE ^ rH rH rH rH rH rH dododr^d^^r^j^j^j^^^dododoor^

£ ~ O «-.fNOfNO^OOOfNOOOOrMfN —».OfNOOfNfNOOOfN Ln^hxovoLnoomfNLnooLnLn^-LncxDoorx^foooTHLDOCT^o & E cn Ul rHrHfNm^-^-LOLnLntDi^oocTJCT» 'H rM ro^ini^iD iD oo Q

rMfNfNfNfNfNrNjfMfNfNrMrMfsifNfMfSfM^^^^^^^^^^ ^ X ■*“' -—• ddddddddddddododododooddddddddddd 5 & E. ^ O O O O O O O O O O O O O O O O rH HrHHrHHrHHHHH ^ Q rOOOOOOOOOOOOOOOOOfNfNfNfNfNfNfNfNfNfNfN fN tH rH rH rH rH H rH rH rH H r—I H rH H rH rH PsJ fN fN fN fN fN fN fN fN fN fN oooooooooooooooooooooooooooo CD rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH tH rH rH rH rH rH rH aj oooooooooooooooooooooooooooo fNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfN

oooooooooooooooooooooooooooooooooooooooooooooooooooooooo

-a r^ooooocnoooocnoocncnoooooooooooo 1 LnOOOOtOtDtDtDtOOOOtDOtOtOOOOOCDOOOOOO op rNoddddddddddddddddddd^^^^^td^ o rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH

■a HcnocTiocncTicriCDcncTicnocnoocnoooooooooooooooooooooo r^r^r^r^r^r^r-f^r^r^r^r^r^f^r^r^r^qqqqooooooo B3 rmmmmdmdmmddmmmrfirrimrrimmfnrrirrirriromm TO I I I I I I I I I I I I I I I I I I I I I I I I I I I I

fNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfN OOOOOOOOOOOOOOOOOOOOOOOOOOOO T 8>

3 in in rH m fN rH rH rH m CD LD CD cd 3 q ID LD LD 00 00 o CD rsj CD 00 o g CD CD rH LD O rsi & LD LD LD 00 fN rH rH rH rH d rH n S LD rn rn rn LD CD rH rH CD rsi rsi rsi rsi rsi rsi rsi rsi d r4 d 4 rH rH rH rH fN g -

LD LD 00 CD CD rsi 00 r>j 00 oo CD 00 fN 00 s 3 g s O q fN g CD rH g 00 CD So 2 00 £ £ S rsi rH d < r rH r-i rsi rsj 00 d rsi rsi rsi rsi rsi rsi rsi r-i rsi rn 4 rsj m d d rH rsj m m LD S ^

jo CD 00 rH O O O CN CD LD LD o o fN O O O LD 00 o oo o f CD LD CD 2 m LD CD OO rsj o 3 o rH r-i d d O O rH rH rH r-i r-i r-i r-i r-i rsi O O d O d O O d d rH’ rsi fN d 3

LD rsi rn 00 LD CD 00 00 LD CD 3 CD CD o m rsj CD LD o o o CD cn g 3 m rH LD g rsi 55 O s 5 - r4 d-i c d CD LO d d d rn d 00 CD 2 d d rsi 2 E fN fN LO d r< rH rH rH rH rsi rsi rsj rsj rn d d d O d d d rH rH m 00 d 3

x T oo q q rH (N 00 rH iq q q CD 00 o oo q q q m q CD ^ H £ d rH rH rsj rH rH rH rH d d d d d rH d rn rn rn rsj rn rsi rn d d Cp rn

O CD O 1^ 00 g o CD CD CD o ID 00 rsj m rsj rsj fN LD 55 m o O O s rsj m rsi rsi g rsi 3 3 rsj rsj O m rn o s äS d o t o n — d d O O o d d d o d o d O O d d d d O O d d d 9 Ö

00 CD CD cd CD m CD m LD 00 g oo rH CD 00 00 00 £-7 o CD CD CD CD 5 s oo 00 CD o o o o o o CD s m rsi ddd ddddddd r4 r4 r4 d d d d d d d CD CD CD r4 d rsi fNI rsi rsj rsi rsi rsj rsj rsi rsj rsi rsi rsj rsi rsj rsi rsj rsj rsi rsj rsi rsi rsi rsi rsi rsj rsi rsi

oo CD 00 LD CD rH rsj m m o 00 rsi fN m CD LD CD £< LO LO Is* <3- 3 00 § CD CD CD S CD g oo o LD c ; rsi 00 LD d rn rsj rsj rsi rn on H d rsi O d r-i d rn r4 d d d r4 CD CD O rsi rn r4 rsi II CD CD CD LD m LD LD ID m rH rsj rsj rsj CM fN rsi rsi m LD LD LD m 0 3 g 2 rsj rsj rsj rsi rsi rsj rsi rsi rsj rsi S 2 2 rsj rsi rsj rsj rsj rsj rsi fN fN rsi rH fN

>■ 00 00 r>- LD ID rsi CD 00 rn rn CT) rsj oo LD rH 2 CD rH 1 5 rH O 00 5 CD S LD g g SSSS LD CD CD cd ^r d d d d <3- d d d d d d d d d m m m m m m rö1/1 m m m m m m m m m m m m m m m m m m m m m m

a i co rq q q q rn rn CN fN 00 fN u 00 m o CD CD r^* CD q q rH ® rsi o r4 CD E d LO rH rH CD CD 00 oo d d d CD (U TO SS S S 2 2 s rH ^ 2

5 ~ rsi rsi 00 o O O o 00 00 00 00 00 rsj VO o VO VO g fN CD CD ° LD r"* CD s LD g CD CD g CD g 3 oo rsi 6 E 00 rH CM m 9 ^r LD CD r^ 00 CD CD rH rsi m 00 CD 0 2

rsj rsj CD CD CD CD CD CD CD CD CD CD CD d d O o O O o o o OO O o OO rsj rsi rsj rsi rsi rsi rsj rsi fN fN fN g g rH t—i O o O O o o o O o o o O o o o O o O O o o o O O 1 & E (N (N rH rH ^ Q rsi rsi s a

O O O o O O o o o O o o o O o o o O O O O o O o O o o o TO rH rH rH rH QJ o o o o § § § g rsi rsj § g § § § § § § § § § § § § g g § g § g fN rsi

00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

LO LD ID LD LD LD LD LD LD LD LD LD LD rH rH rH rH rH rH rH rH rH rH rH rH r** g g m m m m m m m m m m m m m LOd d d dddd ddd dd <3- H rH rH rH rH rH rH rH rH rH rH rH rH rH rH 2 rH rH

oo oo rH rH CT) CT) O) cd 3 3 m m 3 3 s m Jo s s? 5 5 5 5 5 LD rn rr5 rn rn rn rn rn rn rn rn rn rn rn rn rn r-i 2 r-i 2 2 2 r-i r-i r-i 2 r-i r-i rsi ^r rn rn rn rn rn rn rn rn rn rn rn rn rn ? ?

LD LD LD ID LD LD LD LD LD LD LD LD ID 3 4 3 3 s 3 3 3 3 3 3 3 3 8 6

rsj rsi VO VO VO VO CD CD o o o O O O O o o O O O O o O O O o O O O O O O ...... >i >i >. >i >i >i >i >i >i >i >, >, >i oooooooooooooooooooooooooooo rH rH r—I rH rH rH rH rH r-H rH rH rH rH rH rH rH rH rH rH r-I rH rH rH rH rH rH rH rH OOOOOOOOOOOOOOOOOOOOOOOOOOOO s O fNlfNf'JfNfNJfNIfNfNOJfNfMfNfNOJfNfNfNJfNfNfNfNfSfNfNfNrMfNfN in LOLOLOLOLOtOLOCOLDmcOCOLOLOLOLOC/^COLOLOLOLOLOLOCOCOLOLn LOLni/^cnLOLOCOLOLOCOLOCOLOLOLOLOLOLOLOLOCnLOLOLOLOLOLOLO *_ 'o 5 S ^ E d d r^rsi^onoo^^no 3

ocrtm rN r^oi]!-^ •j: % vdlo ^fNHinin^-.. 1/1 E CNifNi fNfN(NHrvil/i^[^ 3

00 00 OOCTiO^OCDrorN 0*1 O O oo

o * l 83 38S3SSS3 2 E d o ddd^°7;3^ 3

ai^inN rninro'tN ^oi r2o 3 rnrorornrornfsindO?

i^OllDHrOCOlD^SSJ 00° 3" rH fN m m fN fN O t: rr!^5!rr' CO -—- dddddddo???

> r ^ r cnroa^^fN cn^cjiroa^m *üj \ fN'tr^^DNCO ^^CO fNrO L/ii/i^Lnini/idddN ^ 1S fNlfMfNfNfNfMfMfNfNfNIM

LocnroT-HOrHLnooocricri S i rNO^roin^rniDkorn^ §2 ° rnrnLndLOLnfNrHr^fNfv! x E m m m m m m ^-Ln^j-oovo o 3 fNfNfNfNfMfNfMfNfNHH

>* fN O O ^^O H inO lH ^fN lO ^DtOr^l^OOOOtNO^U)^^ .E ro m m m m m rom m onfY -jro CO

a] ai oo o on r-: n « h rr n o. != fT E 0) ro S 2 2 S 3 S 2 > ^

£ ~ & E. fN^ri^cr>H H fN jro ^ .ooc^ Q

mrommrommrommro ocr>cr>cr>cr>cr>ocr>cncr>

(m) cnooooocnooocri Max. Depth

ooooooooooo rö rH rH rH rH rH rH r—1 rH rH rH rH OJ OOOOOOOOOOO (NfNfNfNfNfNfNfNfNfMfN

§ 00 00 00 00 00 00 00 00 00 00 00

■a ^ ^ ^ ^ <3; "3;

'5p rH rH rH rH rH r—1 rH rH rH rH rH o rH rH rH rH rH rH rH rH rH rH rH

_aj

3 LD ld lo ltj lo lo lo ld lo LT) lo fNirsirvifsifNfNfNrvirsifNrsi £ mmmmmrommmmm CO i i i i i i i i i i i

§ ooooooooooo 'S oooooooooooooooooooooo 5?

ooooooooooo ooooooooooo

37 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 >1 > ooooooooooo rH rH rH rH rH rH rH rH rH rH rH OOOOOOOOOOO fNlfMfMfNfNfNlfMfNfNfNfN r- LOLOCOLOLOLOLncnLOC/^LO i n COLOLDLOLOLOCOCnLOLOC/l Appendix B - In-situ temperature and potential density

By Laura Richardson and Charles James

The in-situ density of seawater is determined by the salinity, temperature, and pressure. The salinity effect is basically chemistry, the temperature effect is thermodynamic, and the pressure effect is physics with the pressure force moving the water molecules closer together (compressibility of water).

Unfortunately the in-situ density is not useful in situations where static stability is a consideration; it is a circular problem where displacing a parcel of water vertically will change its density in ways that in-situ density can't describe.

The static stability of a parcel of water with respect to the parcel above it or below it has to take into account two effects. The first is the physical effect of compression (all things being equal the parcel becomes less dense if you move it upwards). The second is the thermodynamic effect of compression, where the parcel expands as it cools down (the adiabatic effect).

The dominant effect is actually the compression (despite the fact that we often treat water as incompressible) so the first order correction is to remove pressure from the density calculation and calculate density with p=0 in the equation of state (EOS). Let’s call this sigmaO. Now if the density of a parcel of water measured in sigma 0 is higher than the parcel above it, it will stay denser when displaced upward and try to return to the level it was at, in other words it is statically stable.

The next order of correction is to use potential temperature instead of in-situ temperature. Potential temperature corrects for the adiabatic cooling and heating of a water parcel undergoing vertical displacement. In the previous example the water parcel would remain statically stable as the adiabatic cooling would actually make the parcel of water slightly denser. Potential temperature is defined relative to a pressure surface (most often pref=0) and gives the temperature of the water parcel if it were moved vertically from in-situ to the p=pref level. We often ignore this correction in coastal regions where depths are quite shallow but in very deep profiles ignoring this effect can lead to erroneously unstable profiles.

158 Density calculated with both the compression effect and adiabatic effect (using potential temperature) removed is generally called potential density or in sigma units: sigma theta.

The compression correction reduces the static stability of a density profile while the adiabatic effect increases it. A profile may appear border-line stable or unstable with sigma 0 where it is actually stable with sigmatheta.

Using examples from thesis data

Where:

the temperature is in-situ and the range is ~1 to 22°C

the salinity range is 34.25 to 36 psu

the density is sigma 0 and the range is 25 to 28 kg/m3

the depth range is 0-1200 m (pressure -0-1200 db)

Temperature vs. Potential Temperature

In Matlab the potential temperature can be calculated using a CSIRO Seawater toolbox script:

sw_ptmp(S, T. P, Pref)

A water parcel with in-situ temperature (T) of 1.00°C, salinity (S) of 34.50 and pressure (P) of 1200 db has a potential temperature (theta) with respect to the surface (Pref=0) of

• theta=sw_ptmp(34.5,1,1200,0)= 0.94°C

a difference of 0.06 C

For T=20,S=36,P=200 (typical values from plots nearer the surface at 200m)

• theta = sw_ptmp(36,20, 200,0)=19.96°C

159 a difference of 0.04 C

Errors should be calculated with respect to the expected range of temperatures, which in this case is ~20°C so for the deepest water parcels the % error is -0.3%. Nearer the surface this drops to -0.2% (by definition the error is ~ 0% at the surface).

Density vs. Potential Density ln Matlab the routine

sw_dens0(S, T) calculates density with P automatically set to 0. If we replace T in the calculation with sw_ptmp(S,T,P,0) then we calculate potential density referenced to the surface.

Using the values from the previous example:

• sigma 0=sw_dens0(34.5,1)-1000=27.6434 kg/m' • sigma%heta=sw_dens0(34.5,sw_ptmp(34.5,l ,1200,0))-1000=27.6476 kg/m3

a difference of 0.004 kg/m' and

• sigma_0=sw_dens0(36,20)-1000 = 25.5248 kg/m3 • sigma theta=sw dens0(36,sw_ptmp(36,20,200,0))-1000 = 25.5346 kg/nr

a difference of 0.010 kg/m ’

The range of sigma values in these plots is 25-28 kg/m', about 3 kg/m'. Therefore, for the deepest water parcel the % error is 0.1% and nearer the surface it is 0.3%.

Therefore, the use of in-situ temperature and sigmaO rather than sigma theta in analysis and plotting of the data in this thesis would not show any significant difference, and would not change any of the conclusions made.

160 Appendix C - SAIMOS data file for the Lincoln Shelf (used in Chapters 3 and 4) H 0\ «H p “ 3 op / E 1/1 o O - E Q- co "g X o O E 2 .E CO 1 1 Q a 1 £ a E Q “D QJ C r* . Q H cD CD 03 3 — 03 >* E & •■p 2 4-J sr n ro X — QJ 01 TO § o 03 § | T 3 3 — — 3 3 3 " —' d O rsj d LD CD LT) d d d LO 03 CD s rsi rsi d 3 d d d d m m l LO LO 00 oo oo or oro ro ro ro g g d ro OLO LO 3 /3 in 110 03 m ro Ö d d fN 3 00 CD LO LO g d d d d r^ in 1 1 000 00 ro d d 3 2 110 tH tH d 3 CD CO 5 rsi Q CD fN d or ro ro ro ro rsi 03 d ro 3 110 7\ rsj O ro tH tH r-H O O ro 00 rsi s tH O') LO d d d d LO § § 1 1 1 § 1 00 CN cn m d d 3 110 o d d CD d CT) d ro CD rsi NfN fN LO in d 00 00 OLD LO 00 00 2 ro fN d d d d d d d d d oro ro rH d d CD 00 108 3 ro O d d d d 3 oo d d d d 03 (N d ro > / 00 1/3 r>* 03 00 000 00 00 00 ro rsi ro tH X CD oo 108 HrH rH g g tH d d d 3 rH s s rsi rsi rsi d d 3 3 d oro ro 03 ro 8 rn LT) 00 d m o 127 g LO rH d ro d oo a 00 oro ro 3 3 r^* d d rsi CD 118 3 fN O 00 rsi 000 00 d LD ST d d d d 03 d d 00 00 CD rsi 118 107 d d cn LO d o d rsi rri rsi d O O O O d O d d O rsj LD rsi o l d d ro cn cn in m us oo 00 00 o0 oo 00 oo 00 00 00 00 ro g 5 ro d d d CD ro 100 cn 00 d d d d d o d d d d d rn nro rn LO s s rsi rsi rsi un m or ro ro ro 3 d d d d g 00 § or ro ro ro CD 03 ro 3 3 100 t O ro rH HrH tH d d d d d d LD Ln d in O H oo § 1 or or ro ro ro ro ro CT) O X 100 rH rH g g ro fN CT) d d CD 3 O CT3 00 O ro tH tH tH d in CO d tH Drsi LD CT3 d rsj o03 ro in fN ro tH tH § oo 1 I cn 00 ro d rH ro tH tH d d CD m d 3 oo orri ro m rsj in rsi 03 d m ro ro CD LO fN O Oddd d d LO d ro LO 00 rsi CD 1/3 § 5 ro tH d 03 in d 5 NfN fN 3 d oro ro 3 ro 03 S ooo oo 3 g LO m rn LO in S oo LO O ro ro g O d d d g ro 00 d d 03 03 g d d ro g g 1 tH tH 2 oro ro cn s d d d 110 CD tH d d r-* tH in ro s rsi rsi CT) m O fN g oo tH tH 3 or or or ro ro ro ro ro ro ro in 110 T—1 in LO cn LO St oo d 3 d ro ro d d d CT) § 1 g 1 § 1 tH ro tH in s 110 100 m d d ro tH d d o ro ro LO ro CD O ro 00 NfN fN d d d d m CD CD in tH rsi d d ro ro d d tH X CO o o o 107 3 tH ro o CD d O (N P; d ro r^. g oo rH ro rsi or ro ro ro HtH tH X CO 107 tH nm in 2 d 03 CD d fN CD d CT) 03 rsi d d d m d CD tH tH o rsi d d d ro d d d X 107 100 d HtH tH d d 3 rH fN d d ro fN 00 00 tH tH uo o HtH tH 3 g cn 109 d LO rri in tH O Ln ro rsi CO m oo rsi ro g 3 109 m ro ID 00 m 00 00 ro ro rQ ro rQ ro LO in LO uo uo uo uo rsi tH m LO 00 ID rsj m s 00 in m m m ro d tH d tH d d d tH d d rH d rH r-i d rH O 3 tH t-i t-i 7 \ O t-i 3

fN ro rn ro cn tH ro 00 LD rn tH tH tH tH £ ro o fN 2 m o rsi rsi o O o ro o % d d d d Q-° E9 d O dOOdddOddddd O d d O d d O 3

x Sr- rn O') rn lO CTi cn CT) rn cn rsi in in r—1 in uo 0 o m o o m rsj ro £ o o £ o 00 o ro o o o 00 rsi 2 E rsi d d d T-i rsi d t-i rH d d t-i t—i d 3

1 ~ LD vo tH LO U3 * rsi sP m ro ro rn rsi rn ro rsi ro rsi ro rn ro rsi rri rsi rsj rsj rsi rsj tH rri ro

rn rsi rsi 00 cn tH m rsi 00 cn 00 O — LO LO ro LD ro 2 LD S £ rsi m £ ro 3 rsj rsi LO & m m rsi rsi S rsj rsj 2 S5 d O O to - O d d O d d O d O d d d d d d O O d d d d d O

rsi 00 tH 00 JN UO 00 cn QJ rsj rsi *V r^i CT) 2

rsi rsj ro 00 uo uo ro cn 00 00 cn 00 LO LO LO r^ rH LD cn LD LD l d s £ in m 00 LO LO 00 cn uo Sj ro rsi LD 00 LO d d d d d d d d d d d d d d d d d d d d d d d d d d m m m m m m ro ro m ro m ro ro ro ro ro ro ro ro ro m m m m

QJ 3 rsj rsj in ro tH ro rsi uo uo fN ro tH cn rsi r>- rsj rsj TO „ 00 LO LD LD 53 00 LTj ro 00 uo 00 d d 00 00 00 00 d d 00 d d 00 ro 00 ro oo d Q. “ tH tH

3 ~ LO LOLO LO LTj LO LTj Ln in uo in Ln uo O uo UO in o LO o uo uo uo CT) 00 uo s & £ LO 105 LO LO 110 140 LD Q 3 Q ? X — 2222222SSSSSSSSSSSS53"2»222

cncncncncncncncncncncncncncncncncncncncncncncncncncncn ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo fN(NfNfNrMfNfMfNfNfN(NfNfNfNrNfNrMfNfNfNfMfMfNJfNJrMfNfNl

_0J ^■cpicricri^-^-^j-r^r^r^cDa^cDHHH^-^-rrHT-iHCDcrjCDCDcn ^(NfNiMfNi(NfNqooiDy5^rnrofno^o)0)^^isi/)Lni/)ini/] ujixi^dd^D^jkDi£5

Lr'fSfSrnrY^fY^fY?Lr' L/^ Lr' L/,' L/,i Lr' L/,' u^ L/,i ^ ^ ^ ' ^ v£^ ^ ' ^ ^ ^ ^ ^ m , v V,,, m m m m m m m m m m m m m m m m m m m m ■ I I I I I I I I I I I I I I I I I I I I

T-iT-H<-i(NrMrsioomro^j-»3-

(N vO m m x LO ro ro X 00 ro in X 00 X X X X X X X X X x m x ro ro X If) X ro rsi t-H X X X X X X X X X t-H d t -H d d d d d T-i O T-i ri d T-i t—i d t-i d d d d d rH* rH 43

rsi 5> g g X X t-H CD rsi rn CD X t H X g X m m o o rsi ro rsi o rsi r\i o o o o o o o o O-° 2E d d d d d d d d OO d O d d d O d d d 3

x 3 ID x X t-H ro 00 LO rH X X g g g g LD O o m o o X o o 00 t-H CD rsi o O o rsi rsj o o o o 2 E T-i d rsj ro d d t-H d d d rsi d 3 rt 00 CD CD rsi X CD ro ro r^* X X ro rsj X rn rn r>. <3; X rsj 3 LO rsi rsi T—H t-H ro ro rsi ro ro rsi ro ro rn rsi d d rn <3- d d rn

00 g 00 LO 2 X X ro x rsi rsi X CD t H CD t -H X CD rsi O x rsi s X X t-H X X X X O s s° I O O d O d to —- d O d O d d d O d d d d d d O d d O d d

.-tt fV X cr> (X) 00 LO 00 CD 00 X X 00 t-H CD Qj X rsi X X XX X g t H X X m X CD rsi ro 00 & 00 X z X CD ro X X CD X En c ^ d Lfi d d LOd d d uS d d X X d X X) d X X d X X X d X X X £ § rsi rsi rsi rsi rsi rsj rsi rsi fN rsi rsi rsi rsi rsi rsj rsi rsi rsi rsi rsi rsi rsj rsi rsi rsi

>* x c d rsi ro X rsi 00 ro ro rsi rsi g t-H X X 00 X X X S m F5 CD CD CD CD oo oo X £ X S O X X rsj X 00 X LO Lfi d d d d d d L/i d d d d d d d d d X X X d X X d X X m x m ro x ro ro ro m ro ro ro ro ro x ro X X X XXXXXX X

CD 3 TO ~ t-H X) t -H t -H 00 ro rsj CD X XX CD t -H CD t -H CD fN in f*>s X fc ^ g 00 g x 00 d X 00 00 oo oo d g d d d X d d X d d X d Q . “ rH t -H t -H rsj t -H E QJ

3 - LnLnoSLnooLnooLnLOLnijLnooPiLnoSinoLnLnLnOLnLn & £ Q

S' _ Q E rHfNirNjrslOUOUOÜT-HrHrHT-HTHxHTHT-trHT-HrHTHT-HT-HT-lU’ U’ U1U1U’ x -—

cdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcd ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo rsirsirsirsirsjrsirNrsirsirsirsirsirsicNrsirsirsirsirsjrsjrsjrsirsirsirsirsirsi

t-H t-H t-H tH tH t-H t-H t-H t-h t -H t—iT-Hrsjrsirsir\irsirsirsirsirsir\ir\ir\irsirsi

■a CD^'^-^rxxxrsirNrsixxx^-ri-^cDCDCD'd-^f-^rr^r^r^cDCD Lr)^ !,;t^ !^ 'x!'x!rnrr?nlrH. T-!r1 ^^t ^^rNirNfN rsi rsj o O O X x 1 XXXXXXXXXXXXXXXXXXXXXXXXXXX op xxxxxxxxxxxxxxxxxxxxxxxxxxx t H t H t -H t H t -H t-H t-H t-H t-H t -H t -H t —I r H t -H r H t H t -H rH rH rH rH t-H t-H t-H t-H t -H t-H o

ai 3 mmmrr)mmnnmonmtvfvtVmmmtVii,iirnonor)mm I I I I I I I I I I iii I I I I I

^LDLnLnrxr^r^oooooocriiTio^ OOOOOOOOOOOOO THr-ii-irMrsitNmnnmri-^ cqcücücücococücdcücücücücülocoloooooooooooo XXXXXXXXXXXXXOCOCOCCQCOOOCQCOCDCDCOCC1 CQCO

m o to m to LO m m m m m LO LO LO m CD m LO m LO LO CD m to CM m T—1 o 00 T—1 oo 00 m to to m CD m oo un rsi d tH rsi d 3 d d d d d d d tH d T-i r-i d S r-i d tH d d 3

* ^ cn rn cn rn rn T—1 tH cn o o o o o o 2 o o o o o o o rsj o 2 o T—1 o ^o E£ O d d d d d d d O d d d d d d d 3

x 5> <■7 cn to LO CD cn to to 00 O o rsi o 00 s o s 00 o o o o o m m o cn Gj o tH s CD z E rsi d rsi d rH d rsi d r-i rsi t—i rri d rri rsi rH 3 x ^ 00 rn rn rsi r-. to CD rn rsi rsj O) LO rsj 7 ; LO 7 ; rn 7 rsi PM VP *7 CD to rn ix rn rn ^7 rn rri to rsi <7 rn <7 rri rri rri rri rn rsi rri rri

00 1X1 m to CD rH Qi rsj H oo (7 i LO rsi to t-H P 'S- LO m rsi ^7 m rsi m m s m m 2 rsi m S £ m to £ 2LO z— a? d d d d d d O d d d d d d d O d d d d d d O d d d

00 ■?* *V to LO CT> m rsj m LO rj to to to T—1 7 rn cn UD to LO 00 i/i c m 5 j m m s 2 lO LO rsj m CD m m " m m 2 m 00 to LO to to LO to to LO to to to to LO LO io LO to LO to to LO to to io to to Lfi 1 1 rsi fN rsi rsi rsi rsi rsj rsi rsi rsj rsi rsi rsi rsj rsi rsi rsi rsj rsi rsi rsi rsj rsi CM

—1 7 rn er» T LO 00 to O) 00 LO m j7 rsj 00 LO - LO to to LO O? to n 00 £ to to 2 00 LO 5 LO 00 s Lfi Lfi Lfi LO LO LO LO LO Lfi LO Lfi LO LO LO io LO LO LO LO LO Lfi LO LO LO Lfi LO LO m m m m m m m m m m m m m m m m m m m m m m m m m

QJ 3 03 ^ 00 LO to 00 O) *7 to to T—1 rn *7 to to CO 00 00 7 ; 00 t-H t-H rsi to cn T-H S3 H d LO LO cn to to to <7 LO |7 LO to 7 d LO 7 to LO 00 Q. — s 2 rH 2 2 2 2 rsi T—1 s

3 ~ LO o LO LO o LO LO o LO LO LO LO LO LO LO LO LO LO LO LO

LO to 100 m § § § to cn § 100 to § o 3 Q E X —

moiaimcioimoioioioimoimaimmoioioimtnoimmoiai ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo (NNNtMfM(NfMINNrMNfMfMNfMrMfM(\rMtMtMNNINNNN

(NOI(N(NN(NrMNNN(NfM(MfMNrMN(NN(NrM(NfMNINmm

_0J CT^T-HiHiHrfri-rt^HTHr-taiCT^cri^fLnLnLnrMrMrMLnLnLnaicricri^ iDrnonnncnCTiiTirvrvr^LnLnLn'^-uDtDiDrorooorHTH^-irfr)-^-'^- 1 l/iini/il/ilDlDlDlDlfllfllillilDlDlDlfllDuiuildujlDlDl/il/il/ilD Qj) mmnnmoommrommmmroroonmrooooornromrornmmm O tH t—I t—I tH t—I t—I tH t—I tH t—I tH t—I tH t—I t—I t—I t—I t—I t—I t-HtH t—I t—I t—I tH tH t—I

r^THTHTH^uDkDLnLnLnoooommLnLnLOTHTHT-H^^-^^f^t^-^t ■a rNj000T-iTHrH^^^LnLnrnLnT-HT-|T-H0C)0...T-iT-iTHcx) 3 LOLodLodddddddddLoddLOLOLOtb^^^LOLOLOLO 'S mmmmmmmmmmmmmmmmmmmmv.icommm• I I I I I I I I I I I I I I I I I I I I I I I

rHrHTH(NfNI(NrOPn^- Lnrvrvr^oooooocncncnooo 1/1 1/1 HIo o o o o o o o o OOOOOOOOOO i-HiH tH o o o o m m m rri rri m m CD CO CO CD X X X X X X X xxxxxxxxxxxxxxtr

vO m ro ro LD ro in in in m CD ro r-j ro Ln in in m l o ro ro 5 CD ro rH ro CD ro ro 6 d t-H t-H d O t-H d d t -H *! O d T—H d d T-i O r-i O d t-H O d 3

t H cn CD ro ro g rsj ro £ o ro o rsi rsi o rsj o £ o ro ° 2 o O o O O o o O o Q- E d d d O d d O d d d O d d d O O 3

* 5> ID CD LD no m rsj t-H 00 CD ro 00 cn o o oo 00 ro s rsj s 2 £ s 3 Pl rsj CD ro •3" o rsj m m 2 E d rsi d d d d ro ro d d rsi d d ,-i O O d d O rri ,-i 3

x -s- cn cn CO rsj CD t -H ro t -H t -H r-* ~ > 3 rsi rH rsi rsi ro rsi rsi

O 'S- LO g ro 00 rO Ln rsi rsi S LO P; 3 o LO ' d d d d O d d O d d

ID t H 00 m ro ro rsi m CD 00 cn CD ro CD Ln rsj cn cn ro to in 00 cn 5 cn P i m 3 ro CD cn m 3 cn in s ID LO d LO in d d d d d d d d d d d Lfi d d d d d d d d d d in rsi rsi rsi rsi rsi rsi rsi rsi rsi rsj rsi rsi rsi rsi rsi rsi rsi rsi rsi CN rsj rsi rsi rsi fN Q CM fN

r^cnror^oogchooiDgrn(NiDinrxrnio}^iD^oq[^^j!f5P;j2£^ iniflinininaifli/iinuiini/ii/iinininJDinjGinijiini/iiniflinifl ro m ro m

tu 2 tu .—, uDLn i^»-tr-iuDroijD^3-^-«^-Lf) fNiLn rxono^Ln'3''a’ iD

2 ~ inininrsjooinrsiXinooinooinLnoominrsirMLnLnLninoooin & E. mr^r-H^a^t-HLDriT-HLncrirH^j-r^T-H^-i^T-HLncrit—(^■r^rHUDCTiT-H Q

£

Q E T-_f r—I t__4 — I __ I __I ^—1 — x — t H t-H t-H t-H t-H t H t -H

cncncncncncncncncncncncncncncncncncncncncncncncncncncn ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo rsirsirsirsirsirsjrsirsjrsirsjrsirsirsjrsirsjrsjrsirsjrsirsirsirsirsirsjrsirsirsi

rororororororororororororororororororororororororororo

_0J <3-^-cncncn'd-'d-^'r^r^r^cncncnT-HT-HrHrororoLninLnrsirsirsiLn 1 ^^(NfNfNjfNfNjrNOoqiDiD^romrnLnLn^iDiD^rnrnrnH ^^kD^iii^DUDkDkbkbt£)L/iL/ii/iuiLriL/ikb(i)kbii3^^DLDUDkDii) op mrnmmmmmmroromrommrorommrommmmmmmm o t -H t-H t—1 r H t —I t H t —I t -H t-H t-H t—I tH t—I t-H t-H t—I t-H t—I t-H t—I tH t-H t-H t-H t-H t-H tH

■a — — rrir^.l^ r-mm(r)r~-r^rvrH»HrHr^i^.r^LnLnLntHt-irHQQ OOOO...... ^t’*3**3‘fNfNfNOOOCNfNfNt—It-Ht—(OOO. 2 onmtVrVrViiim m m m m rom m m pnonm m m m m m m rV ■ I I I I I I I I I I I I I I I I I I I

uDUDuarvr^r^oooooocji t-Ht-HtHrsirsirMnompn«^-^3-t^-LnLnLnoooooooooo inmooooooooooooooocacocDcocQcGcDCDcoco ococcocQcocDcocGcocQiDcacQcocococDXXXXXXXXXX

in o m LO LO T—t cn g ro 00 LO LD m LO m CD 00 s o 00 LD r-s oo oo s cn LD 00 CD 3 00 cn 5 00 3 d d d * E d d d d rH d d d d d d t -H d d d d d d d d d d rH d

g H g « ^ o s o o o o o o o o o o o o o o o o o o o o o o 2 E d d d d d d 3

cd g rsi CD LO LD CD ro fN t-H

rsi ro 00 cn un CD ns cn CN LO 00 00 LD o "S' a rsi m rsi LO s ro ro un ro rsi £ rsi ro ro 5 ä? d d OO d d d d d O O d d d O d d d O O LO '

LO T—1 ro LO cn cn LD un rsi LO LD LD LO co CD g If ” e 00 3 00 3 O 3 3 £ 00 00 LD 00 00 cn CD CD CD 3 00 3 00 00 3 LO d LO ld d LO LO d d LO d LQ d CN d d d d d d d d d d LD d fNl rsj 1 1

>* LD l o LO LO ro rsi t-H LO CD CD 00 oo 00 LD cn CN t-H ro cn 00 rsi LO 00 CD ro CD 00 00 00 00 CD So So CD cd CD 00 CD CD So !e LO LO LO LO d LO LO LO LO LO d LO d LO d d d d d d d d d d u-i d d 75 m ro m ro ro ro ro ro ro ro ro m m ro ro ro ro ro ro ro ro m ro ro ro LO aj

3 CD cn CD CD 00 t -H t-H LD rsi rsi ro rsi t -H t -H t -H CD CD CD r-s ro CD 03 _ 00 00 oo g g g g g g 00 00 g oo oo oo 00 00 00 00 00 g g g ns ns g g &ü t-H t-H t -H t -H t -H E

tN^ininQiflO§ino2^o^^ooiflijioino2^o°o

c.

3 RR 2 S 2 3 S 2 2 5 J»«SSSSS 8 S”SSSSS Q E — 1 — 1 —J —J .—I .—I __J ^ T—j ^ .—I LJ 1 SJl LJ! 1 V~M 1 __1 __J _rH 1 —rH 4 __Jt-H __J rv~i X “

cdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcdcd ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo fNfNfN|fMfNJfNfN|(NfNJfNfNfNfNfNfNfNfNfMfNfMfNfN(NfNfNfMfM

fOr0^lDlDlD^lDlD^D^lDU)lDU)'DlD^^DlD^DU)lD'DlD^lD

LnLn^-^-^-cDcncri^'^-^rr^r^r^cncriCDHHHHHHcricriCD^t •a T-HTH^- ^f^fNrNfNrMfNjfNCDCDO'XJUD^Drnn-jror^r^r^unLnu^^t 1 ^D^j^D^i£)ii^D^D^)UDUD>i)i£)(X)i/ii/ii/iLriinini)kDiii>iiaDkDU3 op mmmrommmmmromrorommmmmmrommmmmmm t -H t -H t—I t—I tH t-H t-H t-H t—I t-H t-H t—I t-H t-H t-H t-H t—I t—I t-H t-H t-H t-H t-H t-H t-H t-H t-H o

"D 3 ££ddd££££££dddddddindcocbcbddcdcD ,’ ,’ mni i i o m , , , rVi tyrnmmmmrnromrnnommmmmmI I I I I I I I I I I I I I I I 'S

cn cn in N N m m m i/i oo <-H»-iiHrsirNrMmmmri-^'r}-LnLnLnooooooo cocacncncnooooooooooooooocococococococo XXGCQCQCCQCOCGCÜCQCQCGCGCDCOCOCGCQCÜCQXXXXXXX

sO sO ro rH ro cn LO rH rH CD cn CM ro CD LO oo rH 00 3 LO LD LO £ CD r>. 2 So CD 2 oo s r^ CD s oo cn ro CD CD rsj 00 in " f d d d d rH d O d d d rH d d d d d d O d O rH r-i r-i d d d d 3

5o- ro rH T—1 ro ro CN r\i cn rH o o o O o O o o o o o o o o o o O O rsj rH rH rH OQ- £E d d d d d O O d d d O 3

rsi cd r^ 00 ro CD m CD ro CO cn ro CN CD * 5> rsi rsi 00 ro 00 2 m ro 2 LO 2 £ LO S CD o £ in rH ^r o o d d s d z E O d d r-i O d d d d 3 d d d d d d d ro d d d O rH O 3 rH rsj rH ro ro ro x "S' oo cd ro in rsj CD CD rH ro r'- ro ro ro ro ro ro ro in ro rsj ro rri ro rsi io 3

00 oo CD ro LD ro rH ro rsj 00 3; CD D ro rsi ro ro ro £ ro ro S rsi m ro d s°iO '—'J O d d O O d O d d d d d d O d O d

CD rH rH ro m LO 00 in 00 00 ro rsj (N LD rH CT> o CD CD rsi o rH 2 s 00 rsj 00 00 o 00 s

rsj cn >« 00 CD rsi rM rH 00 00 in CD LO CD 00 rH 00 CD O' 00 LO 00 00 00 S 5} O rsi £ S 00 CD 00 £ LO rsi 00 in CD cri LO LO in CD LO CO CD LO LO CD in in LO in LO in in LO in in in CD in LO Lfi in ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro m m ro

CD 5 ro .— . CD rH r^ rsj rsj rH ro ro CD rsj rH in in ro ro rH 00 LD LO rsj 00 00 00 00 00 CD ro LO LO LO oU 00 00 00 00 00 00 00 00 sQ. ■—- rH rH 2 rH rH E QJ

£ ~ 6 £ Q £ S' „ §§OOO^^^OOo222oOOrNrNfNOOo2222 rv^ rv^ —J --l _, 00 CO 00 ^—| T__f r— r—| .—l.—I.—I.—I l r> l l r\ —^ ,—I ^—1 ^—( —I r-H Q E m m rH t— I rH rH rH r - I L O L O L D tH tH rH rH rH x —

cd cdc dcd cd cdc d cd c d cd c d cd c d cd c d c dc d c dc d c d cd c d cd c d cd c d cd ooooooooooooooooooooooooooo ooooooooooooooooooooooooooo rsirsirsirsirsirsjrsirsirsjrsirsirsirsir\irsirNrsirsirsirsirsirsirsirsirsir\irsi

■a 1 lDujlDlDlDW3lDU)lOujlDujlDlDl/il/il/illi

■a 1 cddcbcbcbdcbcbcbcbcb£$^ninincbiDcDinLninJfi}fiJ5£ '5 rororomrorororororororVi.cnrororororororororVrVI I I I I I I I I I I I I I I I I I I I ■ ■

LO LO CO Q Q Q LnLnu3t£>^or^r''r-'-ooooooCTicncfi00o^53^ OOOOOOOOOOOOOOrHrHrH^*>«> tH rH rH c\l CDCQCQCQCOCQCQCQCQCÜCGCÜCDCÜCQCÜCQ<<

vO r^ m CO IX) CO (O co rsj 00 CO LO COCM CM r-.

rsi t-H CM t-H CO 00 00 00 g rn COCO rH LO <3- (N I—I 00 o o o o CM cm CM o rsi o

00 00 00 co CO X) co 2 oo ud 00 LO o % LO CO s LO LO rsi o o O 00 o CM co o o Pi o 2 E d rri d d d d d d d d d rn d t -H rM d d rsi d CO 3

COm t-H t -H r^» rsi CO 00 lLO o oo co co co fNX "TNP CO «O rn CO CO COCOCO CO

X 00 O) CO rH CO CO <-h r-- s ro CM S rsi CM CO m J s s°co '—- d O O d d O d d d d d d

rH CO rsi LO 00 lo 00 CD CO ±T 'V 5- CO LO LO s X & 00 00 co CO LOPi c \ d co COCO CO CO CO d cb cb cb cb cb LOLOcb cb b cb LO b b b d d 11 rsl rsi fN rsi rsi rsi rsi rsi rsi rsi cm rsi rsi CMrsi rsi CMCMCM rsi

rsi CO CO co LO r-j. CMco co t -H 00 co £ CD 00 CO 00 co 00 00 oo 00 00 co co co Pi S LO co LO LO LOLOLOLOLO x LO LO LO LO LO LOLO LOLO LO LOLO LO LO LO d x m COCOCOco CO co co co COCO COCO COco

QJ B 03 — rsj CO LOLOCOx LOLOr*» co co LOoo t -H rsj COCO rH co LOLO LOLO LOLOLO d LOLOCD co cb 2 CO s b t -H t-H E |QJ

£ - & E 2^25SPffiH|NSHS°rtS°H°°Hl{l0§HSS o

£

Q E —A — I 1 1 —-I —-t _- i 1 1 U 1 —-I —A _ -I ^-1 ^-1 —-J —-I —-I 1 W 1 ^ 1 ^ ^ X “

ocjicricncBcrio^cnCTicnc^^icjio^Gicncrio^^ioi^ic^ffic^c^ ooooooooooooooooooooooooo ooooooooooooooooooooooooo fN|(NfN(NfMfNlfNfNfNlfNllNfNlfNlfNJfNOJfN|(NfMfNfNfNfMfMfN

OOOOOOOOOO^-H^—IrHr—I t-Ht—I t-h r H t-H t -H t-H r H rH t-H t -H r H t-H t -H t-H tH t -H rH rH rH rH rH t-H t-H

-a <3-<3-r^i^cococo<-HrHT-H^-^-^-cococo'3-'3-^-r^r^r^co(Oco 1 rsjrsioococDcocococo^^-^rsirMrNCMrsirsjooocococD

bbbbLnLbLbLnLbi-nbbbbbbbbbbbbLbiOLn 15.31 15.31 11 11 2009 2009 98 98 15 56 16.6 14.1 35.7 35.43 26.1! 26.4! op cococococococococococococococococococococococococococo o r H r H t-H t-H t -H t -H t -H t-H t -H t-H t H t -H t-H t -H t-H t -H t-H t-H t -H t-H t -H t-H t -H t-H t-H t-H t -H

-a n r^n-*j-rMrNirNOOOooooooa!fVrHl: ,: l:rj-'<3-^-rNj

Q_ Q_ Q_ CÜ CD CQ u u u u u u fN CM cm lo lo ld ^-irHTHrsirMrsimonmrt'^r^rLnLn oooo<<<<<

oo m m m LD LO 00 a»m m LDm LO LO LD mCM 00 00 LO s CM £ LO CT CTl CT 81 f d d d d d d d d d lH d d d d d O d d 3

3 . o 00 00 rH CM 00 00 o o o o o o o o o o o o E d d d d d d d d d O d d d 3

3 rsj CN 00 00 VO 00 VD t-H rsi 2 CT00 LO o o o o s o o CTLD o t-H t-H E O t-H d d rH d T-i d T—i r-i d 3 x 'S- CM CM lo CT) t-H rn rn LO CT) LO vq 00 vq CM c- 00 r>* r>. 00 CT) iq CT) CM 00 C". 00 ^ 3 rn rn rn rn rn rsi CM rn CM rn rn rn rn on rn rn CM rn rn d d <3- t-H

m m t o 00 uo LO LO 00 LD 3 00 CT m LD CM LD LO -H cm £ s m CM CM CM & m m m m 5 g 00 LO LD O CMCT si s a? d o LO ' d d d d d O d d d d d d d d d d d d d d d d O d d d

2r rfT ct 00 rN LO rM CM m LO QJ 00 CT) VD rn LD LO LO CT t-H CM CT E S o m 00 rn Si 00 00 m m CM m o CM rn d d d d d d d d d d d d d d d d d d d d d d d d d d d d s 1 CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM CM

rH t-H ct CM CM CM m LO LO LO rn VD 00 vq LD LD rn CT LO rn m 00 LO m lo £ 00 s LO s 3 LD LD s So d d d LOm LOmLO m dmmmmmmm d d d d d d d mmm d d LO mmmmmmmm d d d LD d d d S d mm d d m

CD

503 .—. CT fNJ r>» LO rn LO 00 rn rn 00 00 LO rH 00 VD m LD CM 00 00 LO CT 00 it d d d d 3T LO LO LO CT 00 00 O 00 00 E

£ ~ o o o o 2 LO LO CM LO LO CM 00 LO LO CM LO 00 CM O Si

108 g 105 110 § S LO LD 00 LO CT LO CT 37 4 o £ S' _ Q E X —

ctctctctctctctctctctctctctctctctctctctctctctctctctctctct oooooooooooooooooooooooooooo fNj(Nrsj(NfNrsifNirNjrNjfMfMrMf\ifNirsifMfNjrsirMrsirMojrsirMrsirsirsifNioooooooooooooooooooooooooooo

HfMOJfMfNrMfNIfNfNlfNfMfNfSfNfNfNfNfNfNJfNfNfNfNlfNfNlfNlfNlfN t-H t-H t-H t— I t— I t-H t-H rH tH t— I rH rH rH t-H t-H rH t-H r-H rH t-H t-H t-H t-H t-H t-H t-H t-H t—I

_QJ H^^j-^cna^cn^^^r^r^r^cTicDa^HHHoooocncrj^DUDvD^^ m^^^fNfNfN(N(NfMOOO^iD^rofYiroooooifl^cororn^i^ mrommmmmrommmrommmrommmmmmrommmmm t— I t-H t-H t— I t-H t-H t-H t-H t-H rH t-H t-H t-H t-H t-H t-H tH t-H t-H t-H t-H t-H tH t-H t-H tH t-H t-H

H^i-^r^-_n r n _nr. r. r.fr)fr)rnr^r^r^HHTHCDCT»LnLncr»cr»cnfNrsi Ooooooon! . n! . ,;.^t^-^rNrNirsiooooooorsirsjTHOOOO mmmmrV.....rnmmmmmmmmmrommmmmmm I I I I I I I I I I I I I I I I I I I I I I

lo HHHNiNMnmnTTTinininSSSS H H N (M OLOuicooooooooooooooooggyg C O Q £ 0£ 0£COCOCOCOCOCOCOCOCOCOCOCOCGCOCO^=!^^ Ü u u u

SO rj; m rsi 00 CD QJ PO CD CD 00 rsj rH CD rH OO 00 l d m LD 5 rsi m CD "I d d r-i d d r-i d d r-i d d rH d rH d H m rH rH O fN m LD oo CD m rH m o o o rsi o o 2i d O d d d d d d O d d g rH rsi LD Ol rH rn rH Ox 5~ o o O 00 00 m m 2 o d r-i d d LD d d sd d rn d (N d r-i 2 1 rsj rH O) 00 rsj r^ 00 q 00 rn q LD CD oo LD rsj 00 1^; r- rsj rH LD LD CT) rH -? 3 <3- rn rsi rsj CO rH rn rn rsi rn (N rsi rsi (N rn rn rn rsi fN rsi rsi rsi rsi rsi rsi

P 'S- l d LD m rsi rsi CD rsi l d LD 00 00 00 LD LD LD m m Q- m m fN s £ rsi s fN rsi 3 rsi CO d£ d O d O d d d O d d O d d OO d d d d O d O d d O

rH LD rH oo CD m oo l d 00 m LD LD LD LD .•* •C q LD LD LD LD £ q LD s m 5 m m m LD LD LD d d LO d d LD d dS LD d d LD d LD d LD d d LD d ds LD d d d 1 1 CNI rsj rsj CN rsi rsi rsj fN rsi rsi rsi rsi fN fN rsi fN rsi rsj rsi rsi fN

>* LD Qi rn m o m LD rj- LD 00 fN rH q c? rsj £ rn q S s LD 00 LD s LD rn LD rsi l d LO LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD LD ID LD LD md m md m m m m m m m m m m m m m m m m m m m cu

rH rsj rsj rH rn rn rH 3fT3 „ r^- CD CD rn LD LD LD r- CD LD oo ä oU d rn d rsj d rn d d d LD d rn d fN rsi Q. -—• rsi rH rH E QJ

■£ ~ l d LD m LD o LD LD LD rsi LD LD o rsi LD LD LD

LD 00 CD 106 106 LD 104 108 CD s § s § § 262 Q

5 S' „ Q E —I —I —.J —i _ i T_i<—| T_| T-_|u ’ ' - n U l — x " rH rH rH rH rH rH LD

oooooooooooooooooooooooooo rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH OOOOOOOOOOOOOOOOOOOOOOOOOO (NfNfNrMfNfNfNfNlfNfNfNfMfNfNfNfNINrMfNfNfNirMfSfNfMfNl

rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH

-a ^-<3-

-a qqq^^^rrr^^^fNlNqqqqqcQtxicDqqq1^ 3 oaonro,,,,,,mmmmmmmmonro mnnro,’ £ III I I I I I I I I I I III

> >- >- ooejxxxyyyiu

cocnc/iooooooooooooo<<<<<<<<<< OCnCoecOCOCOCOCOCOCOCOCOCOCOCOGO^UtDUlUOc/lc/lonU o r- \ CD LO CO LO LO rsj rH 00 LD LD ■— o 2 O o LD 2 rsi s rsi LD ro rH ^ E d O r-i d d r H O d d d d d O d d O O 3

cx> LD rsj LO LO a; rH ^r rM 5 o o o o O o LO o o o o ro ro a. E d d d O d O d d d d d d d O 3

LO CD CD Q fN 2 00 r H ro 00 rsi O o o s LD o ro rsi O 00 o o 00 "o 00 CD 2 E d d d d O r-C d O d O rsi LO 3 rsj ro 00 LO LO ro LD CD 00 LD °q 2 ro LD rH 2 LO CD 2 r** CD ro rH LD rio fN rH LO LO LO ro ro rH ro r\i rH rsi fN ro rH ro rsi r H ro rsi rH

(N r-H LO rsi a) LO rH g rsi LD rH oo CD rH LD ro g rsi P 3 rH LD 00 00 Pi 3 ro Pi rH ro 2 ro rsi rsi O S O d O iO —- d d O d O d d d d d O d O d O d O OOOO O

00 CD JO LO LD LD LD LD rH LD 00 ro LO LD LO CD c n LD 00 £ LO ro LD LD LD 2 LD LO 2 3 CD LD d d d d d d d LO LO LO d LO d d LO LO d LO LO d LO d LO d d rsi fN rsi rsi rsi rsi rsi rsi rsi CN rsi rsi rsi rsl rsi rsi rsi fN rsi rsi fN 1 1

> - CD LD LD rsi LD CD rsl LD 00 00 r ^ 00 rH 00 00 00 c d 00 3 S LD LO LO LO 3 LO So LO s c LO LO d d d IT! f S d LO LO LO LO LO LO d LO LO LO LO LO LO LO LO "TO m m ro ro ro m ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro LO QJ 3 LD r^* LD ID LO rsi rsi rsj ro LO 00 rH rsj LO rH LD ro LO TO .—. r H rH CT) 2 o ro o fN O rsi rsi rQ LO fN CD rsj LO rH rsi LO fN LO rsi fN S 2 rH rH 2 rH E O)

O Q j o 2 LO o LO O LO LO LO Q LO o LO § § § 103 105

110 S 362 258 358 (m) Depth

3

S ’ _ _ Q. Q Q 00 00 oo 00 oo oo CD CD cn CD CJl 0") 120 120 101 101 101 109 109 114 114 114 120 109 550 Q E 420 420 X “ TO

o o o O O o O o o o o o o o o o o O o o o o o o o TO QJ S § S § § § § s SS§§§§§§§§ § § § § §

3 r-H r-H rH rH T—1 T—1 rH rsl rsi rsi rsi rsi rsi rsj rsj rsi rsi r\i CN rsi rsi rsi rsi fN rsi §

_CD r H rsj 2 2 00 00 CT> CD 2 2 2 CD CD CD g g g CD CD CD rH =3 00 00 LD LD rsi rsi rsi 3 3 3 LD LD LD ro ro s 4-J d d d d d d d d d d d d d d d d d d d LO LO LO LO LO LO Op rH ro ro ro r o ro r o ro ro ro ro ro ro ro ro ro ro ro ro r o ro ro ro ro o 2 rH r H

c d LO LO m m r H rH rH LD LO i n er» r-. ■S 00 00 fN rsi 3 3 3 ro ro ro £ P p d d d d d d LO LO LO LO LO LO LO LO LO LO LO LO LO LO LO LO LO ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro 'S _1

& O LU LLJ LLJ LO LO § LO 00 00 r-H rH T—1 rH rH rH fN ro g g g s < < OO < < LO LO LO o o OO OO o Co <£> LO LO cd cd cd CD CD CO CO CO § CO CO CO CO CO CO CO CO CO GA19G -36.17 136.67 2 2010 86 15 19.5 36.0! 00 m fN ro LO q ; CD fN ro ro ID ro LO rH LO fN s fN fN LO ro fN LO fN ID s ro £ LO rH in o E T-i rH r-i O d fN d d ro O d r-i d O rH d r-i fN r-i r-i 3

H------fN LO rH ro g 00 rH rO LO rH 5 00 LO O n ro s 0 S O 0 ro O Q_ E O d d d d O d d d d d d d d d O d d d d 3 LO K LO fN LO g LO rsJ 00 fN 00 rH rO 00 ■5 LO O') Ci 0 00 0 ro fN So LO LO 00 ro 0 r-i 9 Z E fN 06 d rH CD d d rH d d d d d LO rH LO d 3 LO fN fN fN ro LO LO ro 00 °o rH fN LO rH LO LO <3 - fN 00 X fN <3 - (N fNJ rH rsi fN fN ro fN r-i rsi rH LO rH fN rsi rsi rH ro rsi fsi IN rn rri

fN ro fN 00 00 00 ro 00 ro 2 LO CD QJ LO r^ ro ro rH 10 rH 2 m fN O LO ro ro Si ro 5 ! O ro fN ro 0 00° d d O "cO £ d d O d O d O O d d O d O d O d d d d O O 9 d O

>* rH 00 VO m rH LO LO

>* LO rH VO LO

5 3 ro 00 LO ro LO ro CTl ID 03 00 fN 00 LO 00 LO 00 CD r*. <- ro fN ro fN fN rH d LO rH d O 00 fN 00 rQ ro O rsi fN 0 rH rH s . £ s E QJ

-5 - SS33asSS2S5S = SS25S§SSS5g2SS & £ a ■5 S' _ Q E »5 SS!lSsnS 2 2 2 - 5 5 SSSSS5 2 2 2 S!; x “

ooooooooooooooooooooooooooo H rH t-H rH rH rH t —I rH t -H rH rH rH rH rH rH rH rH rH rH rH t H rH rH rH rH rH rH ooooooooooooooooooooooooooo (NNN(MrMIN(MNNN(N(MrNNfMN(N(NfMf'INNIN(NNN('l

rsirsirsirsirsirsirsirsirsirsirororororororororororororororororo

_oj !qqS?55ooooSS5?5SSSqqqSSSooooooqo 1 ddddddddddddddddddddddddddd op rorororocororororororororororororororororororororororo rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH o

_0J r-~ ^■ir. ir.,0 CTicnLnLn'3->5i-^j-fr, rr. rr,p n m m r^ r^ r^ m o o m iD L n T-Hix>ooLr!..oooofN rM oooooo...^r^J-^'

S $ £ K £ <§“

<<<<<

tH LD fN ro g tH tH tH ro ro cn fN CTl in ID m o r^ tH 00 O fN fN LD fN fN £ ro rN ro ro fN <3; 2 3 CD 00 on LO d rsi d d d E tH d tH d d d tH rH tH r - i O d r - i 2 d H d d d 3 ~ \ ^ r LO g tH g rH CO tH tH 00 tH tH tH ro CD g o o S o LD fN s O ro o ro O o CO o CL E d d d d d d d O d d d d O d O d d d d d 3 3 \ 00 g ro LD rN LD LD uo ro tH fN ro tH rsi ro cn 00 0X LO ro o cn s fN fN o cn fN o LD ID s oo in o o d Z E d LD d ro d rsi 00 rn ro ro d O r - i O O rt d d 3

G) LD 00 LD ro tH LO fN tH tH LD cn ro LD LD r'- LD G) X 3 oo ro (N tH rv i r i t rn rri CO ro rsi ro rn ro rsi ro rn rn rn rsi rsi fN fN fN rsi H rn

LD tH fN ro g LD 00 LD uo tH LD fN cn ro rn m rN fN LD m 00 cn P 3 ro fN ro 5"£ ro ro fN fN O o ro m m d OO d d d O d d d O d O d O O O d d d d O d O d d

>* tH LD tH l d g tH LD fN CT> ro cn LD ro LD fN LD fN tH g LD s cn cn ro o LD s fN s cn 00 cn s n s ^E d d d c LD LD d LD LD LD d d ID d d d d d LDLD d LD d LD d d d d fN rsi fN fN fN rsi fN (N fN fN fN fN fN fN fN fN fN fN fN fN fN fN fN Q -X

>* ro LD cn LD tH tH LD oo ro cn g g g LD g ro cn cn fN fN 00 cn s LD ro fN fN ro 00 LD cn ro LD S LD ro fN LD ft rn 00 00 cn LD 1 ID LD LD LD d d LD Lfi in LD LD LD Lfi Lfi LD LD LD LD LD LD Lfi Lfi Lfi Lfi Lfi 03 ro ro ro ro ro ro ro ro m m ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro LO QJ 3 LD 00 tH ro ID rsi tH LD r^ rn rn ro rn 00 LD r^- LD ro LD LD LD 03 tH 00 53 tH LD ro ro d 00 00 d LD d ro d d d 00 d LD rn d ro LDLDLD Q- ~ tH tH E QJ

5 - !^tinooofNLnmLounc3oStr!mr^unr-ir}-LnLn2LnLn2inoS 6 1 O

3 S’ _ Q E S§88s^^22a»»SgggSSS8§8SSS X —

ooooooooooooooooooooooooooo tH tH tH tH tH tH tH tH rH tH tH rH tH rH tH t—I tH r-H rH tH tH tH tH rH rH tH tH ooooooooooooooooooooooooooo N(NINNNIMINrMNIN(N(NtNNN(N(MfSfMNNINtN(N(NNN

■a orH^tTHoooooicncri'=r^'d-CTicncntHi-i<-H(Tiocnr^r^r^^trr'53- qmmmoooouifliD't't^iDiDiflmmmiBUwqoo^'l;^ 1 CO LO uS Lfi ID CO CO CD ID CD CD CO ID CD CD LO LO LO LO LO LO i d i d i d i d co cd op mmmmmmmmmmmmmmmmmmmmrommmmmm o T—i t —( t-H t-H t—H t-H t-H t-H rH t H tH t-H t—I t~H rH rH rH t-H rH rH t-H tH t—I rH rH T—I t-H

irtrHTHTHaiCTiLnLnLn^-'^-'a-munLnrHTHrHr^r^r^cnmm^t'^^- -a mOOOM00fMrMNMM00INNNqOON(\|N

CQ CQ CO e? (jj ^ U3U(jgcocoa-a-a- U_ U U U co co co r f m m m oo oo oo oooooocncOLnrslrsitN m m 3 5 5 5 o o^ 2 ^ 3 SS 35 S

m r- LO fN t -H oo rsi LO LO on ro LO ro (N t-H LO 00 00 cn t-H ro CD CD 3 LO 3 S LO ro £ t -H t -H t -H d d d t H H T -i rH O d d d O d d O d o d d “I3

g g 00 t-H t-H h t -H g 5s- O tH O O O t-H O 2 o o o o o o o Q- E d d d d O d d d O d d d d d d d d d d 3

rsi LO fN ro IN ro LO CD LO o f r\j Pi 00 rsi rsi rsi ro ro 2 LO 2 S 2 2 d o d 0.44 0.07 ^ E O d d d O O d t -H d d d d d d d d d 3

X ~ fN LO t-H CD 0 0 ro LO 0 0 t -H 0 0 ro 0 0 LO CD rn ro fN 0 0 ro ro rsi CN NP l O ro ro CO ro mi ro rsi ro rsi rsi t-H rri Lfi Lfi rsi m i rri ro ro

0 0 CD rs j LO CD CD LO t-H rsi ro ro p ~ ro g r ^ £ £ ro s £ ro £ a rsi o LO LO rsi m fN fN ro £ ro s as d l o ■—- d d O d d d O d d O d d O d d O d OOOO d d

Ä 'V q j rH LO ro 0 0 ro CD fN g LO l o ro ro ro t-H ro < 5 ro a ro LO ro 0 0 LO CD d d d d d d d d d d d d d d d d d d I 1 rsi rsi rsi rsi rsi rsi rsi rsi rsi rsi fN rsi rsi fN rsi rN fN fN

CD fN LO g rsi ro t -H CD g CD rO g CD ro t-H g g So 0 0 0 0 s 0? CD CD LO LO CD 0 0 0 0 o O o l o d LO LO io LO l o LO Lfi d l o rn d d d l o Lfi d LO Lfi d d d m m ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro ro

3 ru — t-H t-H CD CD LO LO LO ro CD ro ro 00 ro ro ro LO CD <3; r*. r^ CD fc 0U d LO LO LO LO LO fN t -H LO g g ro LO LO LO d LO LO Q . '— ' t-H td-H d d d d E 0)

3 ~ LO LO LO LO o LO LO 00 LO LO LO LO O QJ o rsi 00 O 00 LO 156 & E 8 § 8 § § 260 364 Q

3 S' _ Q E SS«SSS°°§3;;nSSSa2«»gas X “

oooooooooooooooooooooooot—■I t-H t -H t -H t -H t—I t-H r H r H t-H t-H t-H t-H t -H t-H t -H t~H t-H t-H t-H H t-H r H t-H OOOOOOOOOOOOOOOOOOOOOOOO rsirNir\irvirsjrsi(NfN(NrsjrsirMrNrMrsirsifNfNifNfNfNfNrsjrNj

■a r^r^r^cncncnT-HrHT-Hoooocnoooo^t^-^toocnT-iT-H 1 OOOcocDcDrororooqoqoqcDcDoqoq<3;^J;^rcDcDcqroro dddLOLOLOLOLOLOLOiOLodddddddLOLOLOLOLO W3 mmmmmmmmmmmmmmmmmmrnmmmmrn O tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH tH

mmmr'-r'»r»tHtHiHcncricriLnLncTicn^i-*d-^-r^r'~f^tHtH ■a < Ot-HtHt-|fN fN MDOHOMOONNNOq 3 LO Lfi Lfi Lfi Lfi Lfi LO Lfi Lfi CD CD CD Lfi Lfi *3" Lfi l o l o l o l o Lfi Lfi l o rorororororororororororororororororororororororo 'S I I I I I I I I I I I I I I I I I I I I I I I I

Q-Q-CLCOCOCQ^^^^O Q_ Q_ Q_ CQ CO ^cNojLnLnLni^i^r^oooo n n n in I/i OOO<<<<<<<<<<<33000000<<<<< IAM5CB IAM5CB IAM8SG -35.011AM8SG -35.25 135.31 -35.25 136.69 11 136.69 11 2010 11 2010 2010 95 46 46 80 15 25 15.3 16 16.4 36.08 26.74 36.12 35.77 0.4 26.61 26.24 3.4 0.37 4.4 0.85 0.64 0.11 0.08 CDCDCQoOOOOOOOOOOOOOl/IOOOOOO— lÜCQCQCoOOOOOOOOOOOOOOO

'sf r~ oo oo oo noon_rOOODcni^oocnCTi ,n ^ rH m (N cn cn l/lfNfNor n H - (Y1 ,s ^ 00 0 0 0 0 H rH 'H o *3- <3- l>- ■st o “I odd dd°dddddd<-i d ° r-i d d d d d 3

* 5> oo ro n\i no rH mmiooofNiN'l'ro rsi in in r-x Ln ooooooooooromrorj- o (N 9 o o ^ o o «H °Q- E d d d d d d d d d d d d o d d o d d 3

x f i- m oofnm oiN O i^oo-.in(N (N C Ti m cn O o uor-irMOD'^-rHrvrMO^'^-onLnT-i rH rH o CX) 2 E ddddddddd°n6ono6u-i d d in ° o ni r-i mi 3

x ~ r^ rfUDonnonor-irx'^-ooon^- no no *h ^ UD L/l fN SP to rrirnfnmNniHHninim mi ori ori rsi rn cn

*3" rH LD U0 O r^rsii-HT-HoorfrrJQ^^-iOO oo no no no rxj nororxi»-HO»-iOc-im)';rM N S S (N H ^ ^ s° J LQ '—- d d d d d 0 0 0 0 0 0 0 9 ° ° ° d d d d d ° °

N H H m n O m N10inil),n nOHlD<}INli)H)HHh(N° ° u:>° Lr>Lr'° ° tsnjnin)(M 'N f M N n iN INniNrM nM rM fM rM niniNnininiiN

THr^c»rMannoCTi^<-H^nonr.fHnxoDr^crir^oDcnoDr^rxrHno £ nsioo.oo^HooooCTion^ocxooor^onnooi^i^ooooono idioinjSinioiDininininlilin^inininininininidioinin nonono^’ nononononononom nonononononononononononono

cu a 03 ^ H ri|soO nl ^ r iio rfin ^o |O iiD iI)ni|noni|'} '); fc ou df^ddr^iH iH r^donS^H i-idorioD ^i; noSiDoc)T--iI--iLnno Q. ■—• rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH fN fN rH rH 14.3 35.47 26.41 E QJ

~ £ ______& 1 ?258assss§ssssS2s2ssss;5s Q

a

S' „ Q E ?sssnss22sssa35 = 2s;sss58 x —

OOOOOOOOOOOOOOOrHrHrHrHrHrHrHrHrHrH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH rH ooooooooooooooooooooooooo rsJfNrsIfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNfNCsIfNfNfNfNrsI

■a oi^^^oioioiHHHfnmrorooo^^^aioioiooooisf^ 1 ^^^^^lOOfOforoqqqqH^^^iDiqiooqoqqq dddddddi/iLriiriddddddddddddddd op mmmmmmmrommmmmmmrommmmmmmroro o t-H rH rH rH t—I rH rH rH rH r—I rH t—1 rH rH rH rH rH rH rH rH rH rH rH rH rH

LnTt«3-Ti-cnLninrHrHrHCTia'>cricrir^^i- ^i- *3’ incncn

(J 13

SSSSSSSSSSsasSoSSS^SSSBoS^^ 1AM2CP .AM2CP -35.27 -35.27 135.69 135.69 2 2 2011 2011 99 99 15 75 19.3 35.92 25.6i

on o n m t o ON 4 d d rH u : ID ON q O r H gi d do z E ! N ° d H 3

rHOOlDUDOOrH^J- 3 rri rri rsi rsi r-i rvi r-i

OJ ^ H 00 N ^ on oo on on O O H iO° '—i d d d d d d °

£ 00 Is- 1^ 00 ' f iO ^ CTi no >d; ld od c \ d id d d d d d s I I N ( N ( N ( N (NI PM N

in ^ in oo id >} ct i 00 D in fM N 00 i/i ui i/i i/i oi i/i oo oo oo oo oo oo oo

0) 5 03 _ oo 00 LD 00 rH l d 53 ^ K in oo rsi d Q . '— ’ rH rH rH rH rH t—I rH E QJ

£ - g a $ g s s s 6 E CO rH U3 ^ ^ Q

5

S' _ 5 W 00 00 o o g ° E x — U1 1 ' ' LD ID LO

rH rH rH rH rH rH rH rH rH rH rH rH O O O O O O rsi rsi (N rsi c m rsi

fNJ fN fN fM fN (N fN

_0J CD rH rH rH LD LO LD 0 ro ro rn °o °o °o 1 Lfi LO LO Lfi LO LO Lfi M m m m m m m m O rH rH rH rH rH rH rH

_0J rH rH rH CD CD CD CN O O O rH rH rH

3 l o l o l o l o l d c d o rn rn rn rn rn rn rn 'S

Q_ 1/3 LO LO S g g g g

5 § § S S S 5

sO r-