University of Pennsylvania ScholarlyCommons

Publicly Accessible Penn Dissertations

2014

Diatom-Based Reconstructions of Megathrust Earthquake Deformation and Tsunami Inundation in Central and South-Central Chile

Cristina Dura University of Pennsylvania, [email protected]

Follow this and additional works at: https://repository.upenn.edu/edissertations

Part of the Ecology and Evolutionary Biology Commons, Geology Commons, and the Paleontology Commons

Recommended Citation Dura, Cristina, "Diatom-Based Reconstructions of Megathrust Earthquake Deformation and Tsunami Inundation in Central and South-Central Chile" (2014). Publicly Accessible Penn Dissertations. 1265. https://repository.upenn.edu/edissertations/1265

This paper is posted at ScholarlyCommons. https://repository.upenn.edu/edissertations/1265 For more information, please contact [email protected]. Diatom-Based Reconstructions of Megathrust Earthquake Deformation and Tsunami Inundation in Central and South-Central Chile

Abstract The timespan of instrumental and historical records of earthquakes and tsunamis limit our understanding of long-term subduction zone behavior. Coastal stratigraphy, incorporating diatom sea-level indicators, has provided some of the most detailed geological reconstructions of earthquake induced land-level change and tsunami inundation, often over multiple seismic cycles.

In central Chile, I extended the seismic history through a study of a lowland stratigraphic sequence at Quintero (32.5°S). I documented six laterally continuous sand beds dated to 6200, 5600, 5000, 4400, 3800, and 3700 cal yr BP, probably deposited by high tsunamis. Sediment properties and diatom assemblages of the sand beds--anomalous marine planktonic diatoms and upward fining of silt-sized diatom valves and sediments--point to a marine sediment source and high-energy deposition. I inferred coseismic uplift concurrent with the deposition of the sand beds based on an increase in freshwater siliceous microfossils in units overlying the beds. Our record indicates the recurrence interval of large tsunamigenic earthquakes in central Chile is ~500 years, implying that the frequency of historical earthquakes (~80 year recurrence) in central Chile is not representative of the greatest earthquakes the subduction zone can produce.

My study sites in south-central Chile are located in the overlap of the 1960 (Mw 9.5) Valdivia segment and the 2010 (Mw 8.8) Maule segment ruptures. My paleoseismic investigations from the Tirúa (38.3° S) and Quidico (38.1°S) rivers were consistent with eyewitness accounts of tsunami inundation (AD 2010 and 1960) and historical accounts of coseismic land-level change (AD 2010, 1960, 1835, 1751, and 1575). The vertical deformation inferred from diatom analysis suggests that Maule segment earthquakes result in uplift at our sites (e.g., 2010, 1835, and 1751), and Valdivia segment earthquakes result in no deformation (e.g., 1960) or subsidence (e.g., 1575). I identified four prehistoric tsunami deposits dated to AD 1457-1575, 1443-1547, 256-461, and 176-336. Diatoms indicate uplift coincident with sand deposition in AD 1457-1575 and AD 256-461, which we attribute to the Maule segment. Subsidence coincident with sand deposition in AD 1443-1547, and no change in elevation in AD 176-336, is attributed to the Valdivia segment.

Degree Type Dissertation

Degree Name Doctor of Philosophy (PhD)

Graduate Group Earth & Environmental Science

First Advisor Benjamin P. Horton

Keywords Diatoms, Earthquake, Paleoecology, Paleoseismology, Sea level, Tsunami

Subject Categories Ecology and Evolutionary Biology | Geology | Paleontology This dissertation is available at ScholarlyCommons: https://repository.upenn.edu/edissertations/1265 DIATOM-BASED RECONSTRUCTIONS OF MEGATHRUST EARTHQUAKE DEFORMATION AND TSUNAMI INUNDATION IN CENTRAL AND SOUTH- CENTRAL CHILE

Tina Dura

A DISSERTATION

in

Earth and Environmental Science

Presented to the Faculties of the University of Pennsylvania

in

Partial Fulfillment of the Requirements for the

Degree of Doctor of Philosophy

2014

Supervisor of Dissertation

______Benjamin P. Horton Professor, Rutgers University

Graduate Group Chairperson

______Douglas Jerolmack Associate Professor, Earth and Environmental Science

Dissertation committee

Douglas Jerolmack, Associate Professor, Earth and Environmental Science

Jane K. Willenbring, Assistant Professor, Earth and Environmental Science

Alan R. Nelson (external), Project Chief, USGS

DIATOM-BASED RECONSTRUCTIONS OF MEGATHRUST EARTHQUAKE DEFORMATION AND TSUNAMI INUNDATION IN CENTRAL AND SOUTH- CENTRAL CHILE

© COPYRIGHT

2014

Tina Dura

This work is licensed under the

Creative Commons Attribution-

NonCommercial-ShareAlike 3.0

License

To view a copy of this license, visit http://creativecommons.org/licenses/by-ny-sa/2.0/ DEDICATION

To my Momma. Thank you for making it possible for me to pursue everything I ever

dreamed of. All of this hard work is for you. You are my person for always.

Antes morir que perder la vida.

iii

ACKNOWLEDGMENTS

First I want to sincerely thank my adviser, Ben Horton, the most organized man I know.

His ability to field countless paper and figure drafts and still provide valuable feedback is remarkable. Thank you for always pushing me to do my best, and for being patient through the struggles. I would also like to thank my external supervisor, Alan Nelson.

From our first fieldwork trip to Chirikof, he has gone above and beyond the call of duty to guide me through my scientific endeavors, and I am very grateful. I look forward to many years of fruitful research with both of you.

I also express my gratitude to my committee members, Doug Jerolmack and Jane

Willenbring. Both of them are incredible role models in their passion for science.

I greatly appreciate all the academic and logistical support from the faculty and staff at

Penn. A special thanks goes to Joan and Arlene for their support.

I would like to thank Don Charles at the Academy of Natural Sciences of Drexel

University for providing access to the diatom preparation lab and microscopes there,

Elena Colon for her assistance in the preparation lab, and Mihaela Enache and Sonja

Hausmann for their assistance with diatom analyses.

To the Chirikof crew (Alan, Rich Briggs, Simon Engelhart, Guy Gelfenbaum, and Spoony the fox), thank you for an epic trip and the support and guidance since.

To the Chile crew (Marco Cisternas, Lisa Ely, Rob Wesson, Isabel Hong, Jessica

Pilarczyk, Ed Garrett, Diego Muñoz, Lorena Rebolledo and family, and the students at

Universidad Católica de Valparaíso), thank you for many enjoyable field seasons. I have learned so much from the experience and appreciate your generosity in advice. Marco, I iv am forever grateful for your hospitality and support. Even though you call me “the weed,” I have thoroughly enjoyed working with you. Viva Chile...!

I am indebted to my fellow sea-level research lab-mates, past and present: Simon, Nicole

Khan, Andrea Hawkes, Andy Kemp, and Jessica Pilarczyk. You’ve been invaluable with your eagerness to answer questions and give advice, your support in the field, and importantly, your friendship.

To my fellow grad students (Brandon Hedrick, Kim Miller, Vanessa Boschi, Maddie

Stone, Raleigh Martin, Collin Phillips, and Rachel Valleta), it’s been a pleasure crossing paths with all of you. Thank you for the friendship and support.

To Ivan, I know being in the same house with me while I wrote my dissertation was not the most enjoyable thing you have ever experienced. But, we made it, so now we can survive pretty much anything. I love you, Meat.

To Bodhi, you are my angel starship.

Finally, I want to thank my family and friends in the US, Spain, and Croatia for your support and understanding. I love you guys.

This dissertation would not have been possible without funding from the National

Science Foundation, University of Pennsylvania, National Ocean Science Accelerator

Mass Spectrometer Facility and the Penn Stipend for Summer Paleontology. Lastly, I would like to acknowledge research support from the Greg and Susan Walker

Endowment at the University of Pennsylvania.

No thanks to the wasp.

v

ABSTRACT

DIATOM-BASED RECONSTRUCTIONS OF MEGATHRUST EARTHQUAKE DEFORMATION AND TSUNAMI INUNDATION IN CENTRAL AND SOUTH- CENTRAL CHILE

Tina Dura

Benjamin P. Horton

The timespan of instrumental and historical records of earthquakes and tsunamis limit our understanding of long-term subduction zone behavior. Coastal stratigraphy, incorporating diatom sea-level indicators, has provided some of the most detailed geological reconstructions of earthquake induced land-level change and tsunami inundation, often over multiple seismic cycles.

In central Chile, I extended the seismic history through a study of a lowland stratigraphic sequence at Quintero (32.5S). I documented six laterally continuous sand beds dated to

6200, 5600, 5000, 4400, 3800, and 3700 cal yr BP, probably deposited by high tsunamis. Sediment properties and diatom assemblages of the sand beds—anomalous marine planktonic diatoms and upward fining of silt-sized diatom valves and sediments—point to a marine sediment source and high-energy deposition. I inferred coseismic uplift concurrent with the deposition of the sand beds based on an increase in freshwater siliceous microfossils in units overlying the beds. Our record indicates the recurrence interval of large tsunamigenic earthquakes in central Chile is ~500 years, implying that the frequency of historical earthquakes (~80 year recurrence) in central

Chile is not representative of the greatest earthquakes the subduction zone can produce.

vi

My study sites in south-central Chile are located in the overlap of the 1960 (Mw 9.5)

Valdivia segment and the 2010 (Mw 8.8) Maule segment ruptures. My paleoseismic investigations from the Tirúa (38.3° S) and Quidico (38.1°S) rivers were consistent with eyewitness accounts of tsunami inundation (AD 2010 and 1960) and historical accounts of coseismic land-level change (AD 2010, 1960, 1835, 1751, and 1575). The vertical deformation inferred from diatom analysis suggests that Maule segment earthquakes result in uplift at our sites (e.g., 2010, 1835, and 1751), and Valdivia segment earthquakes result in no deformation (e.g., 1960) or subsidence (e.g., 1575). I identified four prehistoric tsunami deposits dated to AD 1457-1575, 1443-1547, 256-461, and 176-336.

Diatoms indicate uplift coincident with sand deposition in AD 1457-1575 and AD 256-

461, which we attribute to the Maule segment. Subsidence coincident with sand deposition in AD 1443-1547, and no change in elevation in AD 176-336, is attributed to the Valdivia segment.

vii

TABLE OF CONTENTS

ABSTRACT ...... vi

TABLE OF CONTENTS ...... viii

LIST OF TABLES ...... xiii

LIST OF FIGURES ...... xiv

CHAPTER 1. Introduction ...... 1

1.1 Paleoseismic studies on subduction zone coastlines ...... 1

1.1.2 Seismic history of central and south-central Chile ...... 3

1.1.3 The application of diatoms to earthquake and tsunami studies ...... 6

1.2 Thesis aims and objectives ...... 9

1.2.1 Overarching aims ...... 9

1.2.2 Objectives ...... 9

1.3 Thesis Methodology ...... 10

1.3.1 Lithostratigraphy ...... 11

1.3.2 Surveying to sea-level datum ...... 12

1.3.3 Diatom analysis ...... 12

1.3.4 Grain-size analysis ...... 14

1.3.5 Organic content analysis ...... 16

1.3.6 Chronology ...... 16

1.4 Thesis Structure ...... 19

CHAPTER 2. The application of diatoms to earthquake and tsunami studies

...... 21

Abstract ...... 21

viii

2.1 Introduction ...... 22

2.2 Diatoms and coastal environments ...... 25

2.3 Application of diatoms to paleoseismic studies ...... 27

2.3.1 Earthquake deformation cycle ...... 27

2.3.2 Diatoms and relative sea-level changes related to the earthquake deformation

cycle ...... 29

2.4 Application of diatoms to tsunami studies ...... 33

2.4.1 Paleotsunamis ...... 33

2.4.2 Contemporary tsunamis ...... 36

2.5 Concerns and recommendations ...... 38

2.6 Conclusions...... 42

2.7 Acknowledgements ...... 43

CHAPTER 3. Coastal evidence for Holocene subduction-zone earthquakes and tsunamis in central chile ...... 54

Abstract ...... 54

3.1 Introduction ...... 55

3.2 Tectonic and coastal setting ...... 58

3.2.1 Earthquake deformation cycle in central Chile ...... 58

3.2.2 Quintero lowland ...... 60

3.3 Approach and methods ...... 62

3.3.1 Grain-size and loss on ignition analyses ...... 63

3.3.2 Diatom analysis ...... 64

3.3.3 Radiocarbon dating ...... 66

3.4 Results ...... 66

ix

3.4.1 Basal sand ...... 67

3.4.2 Sand bed F and under-and overlying units (Pit 43) ...... 68

3.4.3 Sand bed E and under-and overlying units (Pit 43) ...... 69

3.4.4 Sand bed D and under-and overlying units (Pit 43) ...... 70

3.4.5 Sand bed C and under-and overlying units (Pit 18) ...... 71

3.4.6 Sand bed B and under-and overlying units (Pit 18) ...... 72

3.4.7 Sand bed A and under-and overlying units (Pit 18) ...... 73

3.5 Discussion ...... 74

3.5.1 Evidence for tsunami deposition ...... 75

3.5.2 Evidence for sudden, widespread, and lasting uplift ...... 77

3.5.3 Relative sea level and the limits to preservation of earthquake and tsunami

evidence ...... 80

3.5.4 A 500-year-interval between great earthquakes provides context for the

earthquake record...... 81

3.6 Conclusions ...... 82

3.7 Acknowledgments ...... 84

CHAPTER 4. Late Holocene tsunamis and seismic land-level change in south-central Chile ...... 100

Abstract ...... 100

4.1 Introduction ...... 101

4.2 Background ...... 103

4.2.1 Tectonic setting ...... 103

4.2.2 Tirúa ...... 104

4.2.3 Quidico ...... 105

x

4.3 Approach and Methods ...... 106

4.3.1 Diatom analysis ...... 107

4.3.2 Grain size and loss on ignition analyses ...... 109

4.3.3 Dating ...... 110

4.4 Results ...... 110

4.4.1 Tirúa section 16 (T16) ...... 112

4.4.2 Tirúa section 2 (T2) ...... 119

4.4.3 Quidico section 9 (QU9) ...... 123

4.5 Discussion ...... 128

4.5.1 Evidence for tsunami deposition ...... 129

4.5.2 Evidence for sudden, widespread, and lasting land-level change ...... 131

4.5.3 Earthquakes and land-level change ...... 134

4.5.4 Length of record and recurrence intervals of earthquakes ...... 141

4.6 Conclusions ...... 142

4.7 Acknowledgments ...... 144

CHAPTER 5. Conclusions ...... 163

5.1 Introduction ...... 163

5.2 Summary of thesis ...... 163

5.3 Future considerations ...... 171

APPENDIX ...... 177

A1. Ecology of diagnostic diatom species (exceededing 5% of total valves

counted), central and south-central Chile ...... 177

A2. Full diatom counts from Quintero, Tirúa, and Quidico, Chile ...... 179

BIBLIOGRAPHY ...... 207 xi

INDEX ...... 231

xii

LIST OF TABLES

Table 2.1 Database of noteworthy paleoseismic studies employing diatoms 52-53

Table 3.1 Radiocarbon age determinations, Quintero, Chile 98

Table 3.2 Ecology of diagnostic diatom species in pits 43 and 18, Quintero 99

Table 4.1 Ecology of diagnostic diatom species in sections T16, T2, and Q9 158

Table 4.2 Mean grain size, D10, and LOI of sand beds and units in sections T16, T2, and Q9 159

Table 4.3 Radiocarbon age determinations, Tirúa and Quidico, Chile 160

Table 4.4 Modeled ages for prehistoric sand beds at Tirúa, Chile 161

Table 4.5 DCA axes scores for diatom assemblages within sections T16, T2, and Q9 162

Table A.1 Ecology of diagnostic fossil species, central and south-central Chile 177

Table A.2 Full diatom counts from Quintero, Tirúa, and Quidico, Chile 179

xiii

LIST OF FIGURES

Figure 1.1 Earthquake deformation cycle schematic 3

Figure 1.2 Index maps of central and south-central Chile study sites and historical rupture zones 4

Figure 1.3 Vertical distribution of typical salt marsh diatoms in Chile 7

Figure 1.4 Schematic drawing of coastal uplift/subsidence and diatom response 8

Figure 2.1 Schematic drawing of coastal uplift/subsidence and diatom response 44

Figure 2.2 Summary of paleoseismic studies referenced in the diatom review 45

Figure 2.3 Distribution of modern diatoms in salt marshes in northern Japan and Washington State, USA 46

Figure 2.4 Example of a diatom-based reconstruction of earthquake related subsidence and tsunami inundation in Washington State, USA 47

Figure 2.5 Example of a diatom-based transfer-function approach to reconstruct coseismic subsidence in south-central Alaska 48

Figure 2.6 Example of of a diatom-based transfer-function approach to reconstruct postseismic uplift in Hokkaido, northern Japan 49

Figure 2.7 Diatom analysis of the Storegga Slide tsunami showing fragmentation of diatom valves within the high-energy deposit 50

Figure 2.8 Diatom analyses of the 2004 Indian Ocean tsunami deposit showing grading of diatom valves 51

Figure 3.1 Central Chile index maps including largest historical ruptures 85

Figure 3.2 Schematic drawing of coseismic uplift and tsunami inundation in central Chile 86

Figure 3.3 Location map of the Quintero study area 87

Figure 3.4 Photographs of the Quintero site stratigraphy 88

Figure 3.5 Coast perpendicular profile of the Quintero site stratigraphy leveled to moder mean tide level 89

Figure 3.6 Pit 43 and 18 high resolution grain-size distribution color surface plots with D10 and LOI values and radiocarbon ages shown 90

xiv CHAPTER 1 Figure 3.7 Grain size differential volume curves for stratigraphic units at Quintero 91

Figure 3.8 Summary of microfossil analyses for pit 43 at Quintero 92

Figure 3.9 Summary of microfossil analyses for pit 18 at Quintero 93

Figure 3.10 Results of detrended correspondence analysis (DCA) for diatom assemblages from pits 43 and 18 at Quintero 94

Figure 3.11 Age probability distributions for inferred tsunami-deposited beds from pits 43 and 18 at Quintero 95

Figure 3.12 Summary of evidence for coseismic uplift and tsunami inundation for pit 43 96

Figure 3.13 Summary of evidence for coseismic uplift and tsunami inundation for pit 18 97

Figure 4.1 South-central Chile index maps including largest historical ruptures 145

Figure 4.2 Location map of the Tirúa and Quidico study areas 146

Figure 4.3 Correlated stratigraphic units at Tirúa and Quidico leveled to mean tide level 147

Figure 4.4 Section T16 and T2 high resolution grain-size distribution color surface plots with D10 and LOI values and radiocarbon ages shown 148

Figure 4.5 Section Q9 high resolution grain-size distribution color surface plots with D10 and LOI values and radiocarbon ages shown 149

Figure 4.6 Summary of microfossil and cluster analyses for section T16 150

Figure 4.7 Results of detrended correspondence analysis (DCA) for diatom assemblages from sections T16, T2, and Q9 151

Figure 4.8 Evidence for land-level change and tsunami inundation for section T16 152

Figure 4.9 Summary of microfossil and cluster analyses for section T2 153

Figure 4.10 Evidence for land-level change and tsunami inundation for section T2 154

Figure 4.11 Summary of microfossil and cluster analyses for section Q9 155

Figure 4.12 Evidence for land-level change and tsunami inundation for section Q9 156

Figure 4.13 Summary of historical and paleoseismic evidence for Valdivia and Maule segment earthquakes over the last ~2000 years 157

xv

CHAPTER 1. INTRODUCTION

1.1 PALEOSEISMIC STUDIES ON SUBDUCTION ZONE COASTLINES

The recent Sumatra-Andaman (2004; Mw 9.2) and Tohoku-oki (2011; Mw 9.1) great earthquakes and tsunamis underscored the danger of relying on temporally restricted historical and instrumental records when evaluating seismic hazards at subduction zones

(Subarya et al., 2006; Sieh et al., 2008; Simmons et al., 2011). Prior to the Mw 9.2 rupture of the Andaman-Nicobar segment and subsequent tsunami off of the coast of northern Sumatra in 2004, it was thought that the highly oblique, slow moving segment could not produce a great (>Mw 8.0) earthquake (Stein and Okal, 2007). There were only two documented ruptures along this portion of the subduction zone, in 1881 (M 7.9) and

1941 (M 7.7) (Bilham et al., 2005). Subsequent paleoseismic studies revealed geologic evidence of possible predecessors to the 2004 earthquake and tsunami in northern

Sumatra, Thailand, and the Andaman Islands that could have helped better evaluate the seismic hazards in the region (Jankaew et al., 2008; Monecke et al., 2008; Malik et al.,

2011; Grand Pre et al., 2012). Similarly, the 2011 Tohoku-oki rupture (Mw 9.0) and tsunami off of the eastern coast of Honshu, Japan, exceeded historical fault segment boundaries that were presumed to restrict rupture length and limit earthquake magnitudes to < Mw 8.5 (Earthquake Research Committee, 2005; Simons et al., 2011).

Existing paleoseismic studies outlined extensive, multi-segment prehistoric earthquakes and tsunamis in the Honshu region (Sawai et al., 2008), but only historical earthquakes were included in seismic hazard maps (Hanks et al., 2012; Sawai et al., 2012).

The unexpected magnitude of the 2004 and 2011 earthquakes and tsunamis shows that long-term subduction zone behavior and segmentation can only be fully understood by

1

CHAPTER 1 extending earthquake records over many seismic cycles extending back thousands of years, in order to capture the spatial variability of subduction zone ruptures, and the occurrences of the largest, but least frequent events (Satake and Atwater, 2007). To expand the range of events available for study, paleoseismic investigations employ stratigraphic methods to identify geologic evidence of great subduction-zone earthquakes. Paleoseismic studies focus on coastal sedimentary sequences in low-energy depositional environments such as tidal marshes and estuaries, where distinctive, sharp lithologic contacts signal sudden changes in relative sea level (RSL) related to coseismic vertical deformation of the coastline (Plafker and Savage, 1970; Atwater, 1987). The nature of coseismic vertical deformation accompanying subduction zone ruptures is controlled by the geometry of the megathrust surface and the amount and distribution of slip upon that surface (Fig 1.1; Moreno et al., 2009). Ruptures resulting in coseismic subsidence (RSL rise) along the coast form transgressive (organic-minerogenic) sedimentary sequences (Shennan et al., 1999; Witter et al., 2003), while coseismic uplift

(RSL fall) forms regressive (minerogenic-organic) sequences (Shennan et al., 2009;

Briggs et al., 2014). In addition, coastal sedimentary sequences can record both coseismic subsidence and uplift at one location, depending on the amount and distribution of slip upon the megathrust surface (Briggs et al., 2014; Ely et al., 2014).

Widespread sand beds, deposited by tsunamis accompanying great earthquakes, are often found concomitant with lithologic evidence of sudden RSL change (Sawai, 2001;

Kelsey et al., 2002; Cisternas et al., 2005; Satake and Atwater, 2007). Tsunami deposits often display characteristics such as significant (local and regional) lateral extent, uniform thickness, sharp (1-3 mm) lower and upper contacts, landward thinning, and normal grading (upward fining) of grain size (Dawson and Stewart, 2007; Szczuciński et

2 CHAPTER 1

Figure 1.1: In this example, the coast is in the zone of coseismic subsidence. A coast closer to the trench would experience the opposite deformation cycle. (1) During the interseismic period, strain accumulates between the upper plate and subducting plate causing coastal uplift. (2) When the strain is released during an earthquake, the coast subsides (Modified from Horton et al., 2010).

al., 2012). Paleoseismic methods have been successfully employed to reconstruct earthquake and tsunami records along the coasts of Chile (Atwater et al., 1992; Bartsch-

Winkler and Schmoll, 1993; Cisternas et al., 2005; Nelson et al., 2009), Alaska

(Ovenshine et al., 1976; Combellick and Reger, 1994; Shennan and Hamilton, 2006), and the U.S. Pacific Northwest (Darienzo et al., 1994; Atwater and Hemphill-Haley, 1997;

Nelson et al., 1998).

1.1.2 Seismic history of central and south-central Chile

The 500-year historical record of earthquakes and tsunamis in central and south central

Chile documents a remarkably consistent (80-100 year) recurrence interval of M 8.0-8.5 subduction zone earthquakes and accompanying tsunamis (Fig. 1.2). However, the consistency of the historical recurrence interval is misleading because rupture lengths and slip on the megathrust differ substantially among earthquakes (Moreno et al., 2007;

3 CHAPTER 1

Figure 1.2: Index maps. a) Plate-tectonic setting of Chile in western South America. b) Location of my study areas in central Chile and south-central Chile (shown in red), seismic segments, main tectonic features, and estimated rupture lengths of the largest historical earthquakes in central and south-central Chile since 1570 compiled from Lomnitz, (1970), Kelleher (1972), Comte et al. (1986), Beck et al. (1998), Melnick et al. (2009), Udías et al. (2012), and Melnick et al. (2012).

Cisternas et al., 2012). For example, the central Chile segment has experienced a series of historical subduction zone ruptures (

4 CHAPTER 1 an earthquake or tsunami has not recurred in the last ~300 years, it is not considered to be typical of the central Chilean subduction zone (Vigny et al., 2011). The apparent long aseismic interval between the largest subduction zone earthquakes in central Chile illustrates how temporally restricted earthquake and tsunami records can lead to the underestimation of hazards.

The question of recurrence intervals of unusually large earthquakes and whether segmentation of the Chilean subduction zone is maintained over multiple earthquake cycles also arises in south-central Chile. Two of the largest subduction zone earthquakes in the instrumental record occurred in south-central Chile: the 1960 Mw 9.5 Valdivia segment rupture and the 2010 Mw 8.8 Maule segment rupture. A well-recognized and accepted segment boundary separates the Maule segment and the Valdivia segment at about 38S (Melnick et al., 2009). However, the 2010 rupture called this boundary into question, extending as much as 100 km south into the 1960 rupture area. Furthermore, the distribution of the main slip patches during the 2010 earthquake only partially overlapped with a seismic gap in the Maule segment, possibly increasing the probability of another major earthquake on the Maule segment in the near future (Lay, 2010; Lorito et al., 2011; Moreno et al., 2012).

Despite the questions that remain about how future earthquakes will be arrayed on the

Chilean subduction zone, very few paleoseismic studies employing coastal stratigraphy have been conducted there to lengthen records over multiple earthquake cycles. Even fewer have employed diatoms to reconstruct RSL changes related to coseismic vertical deformation and tsunami inundation (Cisternas et al., 2005; Nelson et al., 2009; Garrett et al., 2013, 2014). Those paleoseismic studies that have been conducted often report

5 CHAPTER 1 fragmentary earthquake and tsunami records that are attributed to net Holocene emergence (Atwater et al., 1992; Nelson et al., 2009). However, Cisternas et al., 2005 identified a more complete stratigraphic record of repeated coseismic subsidence and accompanying tsunami deposits in an estuary near Maullín, verifying the existence of suitable sites for study (Fig. 1.2). Thus, I target similar low-energy estuarine marsh and lagoon environments that have proven effective archives of geologic evidence of past earthquakes and tsunamis, not only in Chile, but in paleoseismic studies around the world (Hemphill-Haley et al., 1995a; Jankaew et al., 2008; Sawai et al., 2009). If optimal geomorphic settings for preservation can be identified in Chile, reconstructions of coseismic vertical deformation and accompanying tsunami inundation associated with prehistoric ruptures can address fundamental questions about the nature of past slip distributions, subduction zone segmentation, and the recurrence intervals of the largest earthquakes and tsunamis in Chile. The results will be relevant not only to the Chilean subduction zone, but also to the Cascadia subduction zone, the Alaska-Aleutian megathrust, the Japan trench, and the Sunda subduction zone, where questions remain about segmentation and recurrence intervals of large earthquakes.

1.1.3 The application of diatoms to earthquake and tsunami studies

Microfossils—such as diatoms—incorporated into coastal sediments have long been recognized as valuable RSL indicators (Fig. 1.3), and provide evidence of earthquake- related RSL change and tsunami inundation inferred from coastal stratigraphy (Fig. 1.4;

Cleve, 1894; 1895; Hemphill-Haley, 1995a). Diatoms are photosynthetic, unicellular algae ranging in size from 5 μm to 200 μm that inhabit freshwater, brackish, and marine environments (Jones, 2007). They are the dominant microphyte in estuarine and marine littoral environments, and their siliceous valves are resistant to taphonomic alteration, 6 CHAPTER 1

Figure 1.3. A vertical distribution of diatom zones (high and low marsh, and tidal flat) typical of salt marshes in Chile.

resulting in their utility in fossil studies (Palmer and Abbott, 1986; Cooper et al., 2010).

Diatoms have been described and classified for over 200 years based on their different shapes, sizes, and the intricate morphological characteristics of their siliceous valves

(Round et al., 1990). Systematic and taxonomic investigations of contemporary and fossil diatoms began to be supplemented by studies of distributional ecology in the late

1890’s (e.g., Cleve, 1894; 1895). Diatoms are a valuable tool in reconstructing paleoenvironmental changes because of their preferences to a number of environmental factors including salinity, tidal exposure, substrate, vegetation, pH, nutrient supply and 7 CHAPTER 1

Figure 1.4: Schematic drawing of coseismic uplift (a) and subsidence (b) and accompanying tsunami inundation. (1) When the strain is released during an earthquake, the coast suddenly subsides (a) or uplifts (b); (2) If the earthquake has produced a tsunami, it will impact the coast minutes to hours after the earthquake; (3) During the interseismic period, strain accumulates between the upper plate and subducting plate causing gradual coastal subsidence (a) or uplift (b).

temperature. Comprehensive ecological classifications such as those by Hustedt (1937,

1939, 1953, 1957), Lowe (1974), Patrick and Williams (1990), Denys (1991-1992), Juggins

(1992), Vos and de Wolf (1988; 1993), and Lange-Bertalot (2000) classified diatom taxa based on how they respond to these environmental factors.

Diatoms’ preferences for salinity and their life form are particularly valuable for earthquake and tsunami studies. Changes in salinity across the intertidal zone produces vertically zoned diatom assemblages with respect to the tidal frame (Fig. 1.3; Hemphill- 8 CHAPTER 1 Haley 1995b; Sherrod et al., 1999; Horton et al., 2010) that can be applied to fossil sediments to reconstruct RSL (Fig. 1.4; e.g., Zong and Tooley, 1996). Diatoms have been applied to historic and prehistoric earthquake and tsunami reconstructions at several subduction zones (Hemphill-Haley, 1995a; Hemphill-Haley, 1996; Shennan et al., 1996;

Sawai, 2001; Sherrod, 2001; Garrett et al., 2013). However, the relation of modern diatoms to environmental factors (salinity, tidal exposure, and substrate) and fossil diatom taxonomy is not well defined in many locations due to a scarcity of diatom data, often complicating interpretations of prehistoric RSL change (Nelson et al., 2009). In addition, the application of diatom biostratigraphy to earthquake and tsunami studies has mostly focused on prehistoric events, making it difficult to test the efficacy of the method (Nelson et al., 2008; Kelsey et al., 2005).

1.2 THESIS AIMS AND OBJECTIVES

1.2.1 Overarching aims

1. Assess the applicability of diatoms to the reconstruction of RSL changes related

to the earthquake deformation cycle and tsunami inundation over a range of

geomorphic environments and temporal scales.

2. Apply diatoms to reconstruct the earthquake and tsunami record of central Chile

and south-central Chile over multiple earthquake cycles (thousands of years) to

address fundamental questions about the slip distributions, segmentation, and

recurrence intervals of the largest subduction zone earthquakes and tsunamis.

1.2.2 Objectives

The objectives defined to address the overarching aims are as follows:

9 CHAPTER 1 1. Review the application of diatoms to paleoseismic studies in order to identify the

advantages and disadvantages of the method, and provide a resource for future

investigations.

2. Identify suitable geomorphic environments in central and south-central Chile

that preserve geologic evidence of past earthquakes and tsunamis for detailed

stratigraphic and biostratigraphic analyses.

3. Characterize site sediments based on their lithology, grain size distribution, and

organic matter content.

4. Develop a taxonomic database of fossil diatoms in central and south-central Chile

and determine the salinity and substrate preference of fossil diatoms from each

site using existing modern diatom studies.

5. Explore intra- and inter-site variability in diatom evidence for RSL change

related to the earthquake deformation cycle and tsunami inundation.

6. Compare diatom-based reconstructions with historic documentation of coseismic

deformation and accompanying tsunamis.

7. Extend the earthquake and tsunami record of central and south-central chile and

develop recurrence intervals of great subduction zone earthquakes.

8. Explore the Holocene RSL history of sites and the limits it imposes on the

preservation of earthquake and tsunami evidence.

1.3 THESIS METHODOLOGY

During 4 field seasons in 2011-2014 I conducted a stratigraphical study in a coastal lagoon and two coastal floodplains in central and south-central Chile, respectively. My stratigraphical investigations revealed sequences of buried soils and sand beds indicative of coseismic land-level change during large earthquakes and tsunami deposition.

10 CHAPTER 1 Sections representative of site stratigraphy were selected for detailed laboratory analysis; monoliths were collected in the field and subsampled for diatom analysis, grain-size distribution, organic content (loss on ignition), and chronological analyses.

1.3.1 Lithostratigraphy

I described cores, hand-dug pits, tidal bank sections, and exposures to depths of ~2-3 m along >200 m stratigraphic transects. Multiple stratigraphic transects were positioned perpendicular and parallel to the coast.

For coring, I used a 1-m-long, 25-mm-diameter, half-cylinder gouge corer. All sections were photographed and their lithology, color (Munsell Soil Color Chart, 1975), and stratigraphy described in the field. The lithostratigraphy of all samples was recorded using the Troels-Smith (1955) scheme of stratigraphic notation. This descriptive scheme allows direct comparisons with data obtained by other researchers (Tooley, 1978). For each stratigraphic unit identified the following results were recorded:

• Upper and lower depths below ground surface;

• Composition of unit;

• Degree of humification of organic horizons;

• Physical properties of the unit;

• A short description.

11 CHAPTER 1 1.3.2 Surveying to sea-level datum

I surveyed all cores, hand-dug pits, tidal bank sections, and exposures using a differential GPS system (Leica 1200). All were tied into local benchmarks and tidal datums and reported as elevation (in meters) relative to mean tide level (MTL). This allows easy compilation of results from different locations, in particular if the coastline has moved a significant distance since the formation of the proxy indicators.

1.3.3 Diatom analysis

Diatom-based reconstructions of relative sea level (RSL) changes related to the earthquake deformation cycle are based on the relationship between diatom assemblages and elevation with respect to the tidal frame (e.g., Zong and Horton, 1998). Over time, diatoms become incorporated in coastal sediments, resulting in buried assemblages that represent an environmental history that can span thousands of years (Witter et al.,

2003). Sudden changes in diatom assemblages across transgressive or regressive contacts can signal an abrupt change in depositional environment from earthquake- related RSL rise (land-level subsidence) or fall (land-level uplift), and can be used to estimate the amount of coastal deformation (Hamilton et al., 2005). Diatoms have been used to reconstruct earthquake and tsunami histories at several subduction zones

(Hemphill-Haley, 1995a; Shennan et al., 1996; Sawai, 2001; Shennan and Hamilton,

2006). The utility of diatoms as RSL indicators also stems from the high preservation potential of their siliceous valves in coastal sedimentary archives (Hamilton et al., 2005;

Sawai et al., 2008).

In fossil sections, I subsampled for diatom analysis at 2 cm intervals near sharp lithologic contacts that may represent sudden changes in RSL related to the earthquake

12 CHAPTER 1 deformation cycle, and 15-cm intervals in homogenous sediment. To remove organic material, I oxidized ~1 g of sediment with nitric acid in a microwave digestion system

(Charles et al., 2002). I then pipetted and evenly distributed a known volume of digested sample (between 25 and 100 l depending on the diatom concentration) on a cover slip. I dried the cover slip overnight and then inverted and mounted it on a glass slide using

Naphrax (Charles et al., 2002). I identified diatoms to species level using a light microscope under oil immersion at 1000x magnification with reference to Krammer and

Lange-Bertalot (1986, 1988, 1991a,b) and Lange-Bertalot (2000). I identified and counted between 300 and 400 diatoms in slides, with each species expressed as a percentage of total diatom valves counted. Fragments containing more than half a valve were included in the count. Only species that exceeded 5% of total valves counted were used for paleoecological interpretations (Horton et al., 2007). I classified diatoms by salinity (marine, brackish, and freshwater) and life-form (planktonic, epipelic, epiphytic, aerophilic) based on the ecological preferences outlined in Chilean (Rivera, 1983;

Rebolledo et al., 2005, 2011) and global (e.g., Krammer and Lange-Bertalot, 1986, 1988,

1991a,b; Denys, 1991; Vos and de Wolf, 1988, 1993; Hartely et al., 1996; Lange-Bertalot,

2000) catalogs. I also scanned samples for phytoliths and chrysophyte cysts to help distinguish freshwater from tidal environments. Phytoliths typical of brackish and fresh coastal grasses were identified to the subfamily level (Pooideae and Bambusoideae subfamilies) with reference to the classifications of Lu and Liu (2003).

I determined the depositional environment (high-marsh/upland, brackish lagoon, low- marsh, or tidal flat) represented by each stratigraphic unit based on a qualitative interpretation of the relative abundances of diatoms within salinity and life-form groups.

The marine diatom group includes marine and marine-brackish species that thrive in

13 CHAPTER 1 salinities exceeding 30 practical salinity units (psu). The brackish diatom group includes brackish-marine and brackish species that tolerate salt concentrations between 0.2 and

30 psu. The freshwater group includes fresh-brackish and fresh diatoms that generally occur in salt concentrations less than 0.2 psu. Diatom taxa that live attached to plants are defined as epiphytic forms; taxa that live on wet sediments are defined as epipelic forms; taxa that live on wet sediments but are able to survive temporarily dry conditions are defined as aerophilic forms. Tychoplanktonic diatoms include an array of species that live in the benthos, but are commonly found in the plankton. Diatoms that float in the water column and do not live attached to any substrate are defined as planktonic forms

(Vos and de Wolf, 1988, 1993).

I defined diatom assemblage zones using a stratigraphically constrained incremental sum-of-squares (CONISS) cluster analysis. I also applied detrended correspondence analysis (DCA) to examine the pattern of diatom assemblage variation within and between samples, and to identify environmental gradients. Higher eigenvalues represent strong environmental gradient, such as salinity, along an axis. Samples with similar species compositions are grouped together in the DCA bi-plot and samples with statistically different species compositions plot apart (e.g., Birks, 1986; 1992). The percent abundance of grass phytoliths and chrysophyte cysts were included in the DCA analysis.

1.3.4 Grain-size analysis

One of the most commonly reported features of tsunami deposits is their sediment grain size characteristics (e.g., Dawson et al., 1991; Minoura and Nakaya, 1991; Gelfenbaum and Jaffe, 2003; Goff et al., 2004; Moore et al., 2006, 2011; Paris et al., 2007; Higman

14 CHAPTER 1 and Bourgeois, 2008; Morton et al., 2008; Fujino et al., 2010). Tsunami deposits are primarily supplied from local sedimentary environments (e.g., soil, sandy beach, dune, and tidal river channel sediments), thus grain-size data makes it possible to interpret the likely sediment source. Grain-size distributions of tsunami deposits also reflect the hydrodynamic conditions at the time of deposition (e.g., Gelfenbaum and Jaffe, 2003;

Moore et al., 2011). The most common parameter mentioned is a landward and upward fining of the deposits that is related to a decrease in water flow velocity and ability to transport sediments, and a shift to sediments settling out of suspension (Dawson and

Shi, 2000).

In fossil sections, I subsampled for grain size at 1 cm intervals near sharp lithologic contacts and within sand beds in order to identify sudden changes in depositional environment and upward fining within the beds, respectively. I subsampled at 5 cm intervals in homogenous sediments. Modern and fossil grain-size samples were treated with hydrogen peroxide (30%) to remove organic material and analyzed using a

Beckman Coulter LS230 laser diffraction grain-size analyzer. Grain-size interpretations are based on the Wentworth Phi Scale and reported as differential volume (the percentage of total volume that each size class occupies). The grain-size distributions were interpolated and gridded using a triangular irregular network (TIN) algorithm

(Sambridge et al., 1995) and plotted as color surface plots using Geosoft Oasis TM software (Donato et al., 2009). I used the mean grain size and D10 value (diameter at which 10% of a sample's volume comprises smaller particles) of each sample to characterize lithologic units.

15 CHAPTER 1 1.3.5 Organic content analysis

Organic matter content provides valuable evidence of coastal change related to the earthquake deformation cycle, much of it as the first or last indications of

“terrestrialization” or evidence of subtle or temporary changes that do not necessarily lead to any large-scale environmental change that is expressed in the litho- or biostratigraphy (Plater et al., 2015). Organic matter further characterizes tsunami sediments and helps distinguish them from underlying and overlying units.

Loss on Ignition (LOI) is usually applied as a measure of the organic content of the material. To measure LOI, I determine the content of inorganic material of dried sediment that remains after being ignited at 550 °C (Aaby 1986). Following the grain size sampling scheme, fossil samples for LOI were dried at 105C for 12 hours, weighed, baked at 550C for 6 hours, and weighed again to estimate the weight lost (LOI; Ball,

1964).

1.3.6 Chronology

Radiocarbon-based chronologies have improved estimates of the timing of great earthquakes (Atwater et al., 1991, 2005; Nelson et al., 1995; Atwater and Hemphill-

Haley, 1997; Cisternas et al., 2005), and tsunamis (Nanayama et al., 2003; Kelsey et al.,

2005; Bourgeois et al., 2006; Sawai et al., 2009) and help establish inter-event times and recurrence intervals.

Radiocarbon (14C) dating has been employed extensively in sea-level research as a means of establishing an absolute age for a sea-level indicator. Libby (1952) was the first to report the principles and methods of 14C dating. He stated that 14C is continually

16 CHAPTER 1 produced in the Earth’s upper atmosphere and is taken up directly by plants during photosynthesis and indirectly by other life forms through the food chain. When the organism dies the carbon uptake ceases and the 14C isotope begins to decay. Therefore, given the standard of 14C activity and the half live (5570 ± 30 14C years) of the isotope, an age can be calculated by measuring the 14C of the sample. The process of measuring 14C activity is described by Bowen (1978). Since the late 1980s, the precision and accuracy of chronologies developed for coastal sedimentary sequences have improved through accelerator mass spectrometer (AMS; Atwater et al., 1991; Törnqvist et al., 1992;

Marshall et al., 2007). Organic and clastic material continuously deposited and buried within low-energy, anaerobic tidal marsh environments for decades to centuries can form sedimentary sequences that are meters thick. AMS radiocarbon dating of individual plant macrofossils within these sedimentary sequences can provide a chronology of marsh sedimentation and the sudden lithologic changes related to the earthquake deformation cycle therein (Warner, 1988; Birks, 2002).

I used radiocarbon and 137Cs analyses to constrain the ages of sand beds in our stratigraphic sections (Ely et al., 1992; Kemp et al., 2013). I collected terrestrial plant macrofossils for radiocarbon dating from organic sediments that gave either limiting maximum ages (usually below each contact) or limiting minimum age (in growth position above each contact) to constrain the age of the sand beds. Following Kemp et al.

(2013), I collected samples that were in growth position (rhizomes) or so delicate (e.g., seeds, leaf parts) that they would probably have decayed or been broken by transport soon after death, thus reducing the likelihood of dating material much older than contact burial. Terrestrial plant macrofossils are strongly preferred as 14C sample material because they have only exchanged carbon with the atmospheric reservoir, thus avoiding

17 CHAPTER 1 the difficulties associated with estimating the degree of 14C depletion in aquatic or marine samples, which typically incorporate carbon derived from a mixture of nonatmospheric reservoirs, such as the ocean, estuaries, and lakes (Törnqvist et al.,

1992; Wohlfarth et al., 1998; Nilsson et al., 2001). Radiocarbon ages were calibrated using OxCal radiocarbon calibration software (Bronk Ramsey, 2009) with the SHCal13 data set of Hogg et al. (2013). Calibrated age ranges are shown at two standard deviations in Years BC/AD or years ‘before present’ (BP; years before AD 1950). I used maximum and minimum limiting radiocarbon ages to calculate age probability distributions for inferred tsunami beds with the V-sequence feature of OxCal (Fig. 3.11;

Bronk Ramsey 2008; e.g., DuRoss et al., 2011; Berryman et al., 2012). To estimate inter- event times and recurrence intervals of paleoearthquakes and tsunamis I used the

“difference” function in OxCal.

Due to age ambiguities caused by the broad plateau in the calibration curve for the past

300 years, dating of younger material can be supplemented by 137Cs analysis. The presence of 137Cs is directly related to the atmospheric testing of nuclear devices during the latter half of the 1950s and early 1960s, thus, its presence in the stratigraphy indicates deposition after the initiation of nuclear testing. Samples for detection of 137Cs were collected from multiple stratigraphic horizons above and below the assumed AD

1960 sand layer at our south-central Chile sites. 137Cs is typically not mobile in sediments containing clay or silt (Ely et al., 1992), rendering them suitable for this method of dating. 137Cs analysis was conducted at Microanalytica Laboratory in Miami, Florida.

Samples with a high organic content were heated to leave only noncombustible ash prior to analysis with a gamma spectrometer.

18 CHAPTER 1 1.4 THESIS STRUCTURE

Chapter Two fills a gap in the microfossil literature as no previous publication has focused on the application of diatoms to paleoseismic studies. The chapter reviews the published literature, elucidates the relationship of diatoms to a variety of environmental variables, and summarizes the advantages and disadvantages to using diatoms as RSL indicators, providing a resource for future investigations. Furthermore, the terminology used in paleoseismic studies and the methods of diatom-based RSL reconstructions are described. This chapter has been prepared to be submitted to Earth Science Reviews.

Chapter Three presents the results of a paleoseismic investigation that employed stratigraphic, lithologic (grain-size and loss on ignition), diatom, and radiocarbon analyses to reconstruct a record of Holocene earthquakes and tsunamis along the central

Chile segment of the Chilean subduction zone. The chapter describes the process of identifying a suitable geomorphic environment that contained geologic evidence of past earthquakes and tsunamis, describes new fossil diatom taxonomy, and applies diatom results to identify prehistoric coseismic land-level change coincident with tsunami deposition. The extended earthquake and tsunami record provides insight into the influence of RSL on the preservation of geologic evidence of past events and the recurrence interval of the largest earthquakes on the central Chile subduction zone. This chapter has been published in Quaternary Science Reviews (Dura, T., Cisternas, C.,

Horton, B.P., Ely, L.L., Nelson, A.R., Wesson, R.L., Pilarczyk, J.E., 2014, Coastal evidence for Holocene subduction-zone earthquakes and tsunamis in central Chile.

Quaternary Science Reviews.

19 CHAPTER 1 Chapter Four presents the results of a paleoseismic investigation that employed stratigraphic, lithologic (grain-size and loss on ignition), diatom, and radiocarbon analyses to reconstruct a record of historical and prehistoric earthquakes and tsunamis at the overlap of the Maule and Valdivia segments of the Chilean subduction zone. The chapter describes new fossil diatom taxonomy, and tests the efficacy of diatom-based reconstructions of RSL change related to known historical coseismic vertical deformation before applying them to the prehistoric record. In addition, the chapter investigates the intra- and inter-site variability of diatom-based reconstructions of RSL change. This chapter has been prepared to be submitted to the Geological Society of

America Bulletin (GSAB).

Chapter Five The final chapter considers to what extent the initial research aims and objectives stated in Section 1.2 have been met and proposes themes for future research.

Appendix 1 Eighty-two diatom species that exceeded 5% of total valves counted are included in a fossil diatom database table, and were used for paleoecological interpretations.

Appendix 2 Full diatom counts for central Chile (Quintero) and south-central Chile

(Tirúa and Quidico) are included.

20

CHAPTER 2. THE APPLICATION OF DIATOMS TO EARTHQUAKE AND TSUNAMI STUDIES*

*To be submitted to Earth Science Reviews as: Dura, T., Hemphill-Haley, E., Sawai, Y., and Horton, B.P., The application of diatoms to earthquake and tsunami studies.

ABSTRACT

Temporally restricted historical and instrumental records limit our understanding of long-term subduction zone behavior; datasets on centennial and millennial temporal scales are necessary to capture the spatial variability of subduction zone ruptures, and the occurrences of large, but infrequent events. Microfossils, such as diatoms, incorporated into coastal sediments provide some of the most detailed geological reconstructions of earthquakes and tsunamis, often over multiple seismic cycles. In this review we explain how the relations between diatoms and salinity, tidal elevation, and life form are used to reconstruct records of vertical land movement associated with large earthquakes, and identify anomalous sand and silt beds deposited by tsunamis along tectonically active coastlines. We compile a database of diatom-based earthquake and tsunami reconstructions from Chile, the Indian Ocean, Japan, New Zealand, the North

Sea, the Pacific Northwest, and the South Pacific. In the Pacific Northwest, diatoms support stratigraphic inferences of coseismic subdsidence associated with a large subduction zone rupture in AD 1700, and ascribe a tidal flat source to a widespread sand bed deposited by a tsunami shortly following the earthquake. In Alaska and Japan, diatom-based transfer functions provide continuous records of vertical land movement during complete earthquake deformation cycles. Analysis of the 2004 tsunami deposit in

Thailand illustrates the typical mixed diatom assemblage found in overwash deposits as a tsunami erodes, transports, and deposits marine, brackish, and freshwater sediments 21

CHAPTER 2 as it inundates coastal and inland areas, and reveals grading of diatom valves within the deposit. Studies in the North Sea use the taphonomic character (e.g., fragmentation) of diatom valves to identify tsunami sediments deposited by the Storegga Slide tsunami.

Finally, we discuss the advantages and disadvantages of diatom-based reconstructions of earthquakes and tsunamis and emphasize the importance of studying the modern diatom response to changes in RSL and the diatom composition and taphonomic character of modern tsunami deposits to improve reconstructions.

2.1 INTRODUCTION

An incomplete understanding of the seismic hazards associated with the Sunda and

Japan trenches contributed to the devastating impacts of the 2004 Indian Ocean and

2011 Tohoku-Oki great earthquakes and tsunamis (Rhodes et al., 2006; Geller, 2011;

Stein and Okal, 2011; Heki, 2011). Instrumental records of previous earthquakes and tsunamis proved too short to estimate the potential magnitude and recurrence interval of these rare events (Stein and Okal, 2007). With more than a third of the world’s coastlines lying adjacent to active plate boundaries (Stewart and Vita-Finzi, 1998), it is imperative to extend earthquake and tsunami histories back in time to properly assess seismic hazards (Small et al., 2000; Satake and Atwater, 2007; Stein and Okal, 2011).

Stratigraphic evidence of great subduction zone earthquakes, first recognized in the coastal wetlands of Alaska (Bartsch-Winkler and Schmoll, 1987; Combellick, 1991, 1994;

Combellick and Reger, 1994), the Pacific Northwest (Atwater, 1987; Darienzo et al., 1994,

Nelson et al., 1996a), Chile (Bourgeois and Reinhart, 1989; Atwater et al., 1992a) and

Japan (Sawai et al., 2002; Nanayama et al., 2003), has been used to extend records over multiple earthquake cycles. Coseismic vertical deformation of the coastline is recorded in

22 CHAPTER 2 coastal sedimentary sequences as a series of distinctive, sharp lithologic contacts that represent sudden changes in relative sea level (RSL) (Nelson et al., 1996b; Yeats et al.,

1997). Widespread sand beds, deposited by tsunamis accompanying great earthquakes, are often found concomitant with lithologic evidence of sudden RSL change. In addition, the largest subduction zone earthquakes can produce trans-oceanic tsunamis that deposit anomalous sediments on coastlines hundreds to thousands of kilometers away from the earthquake source (e.g., Bourgeois et al., 1999; Gelfenbaum and Jaffe, 2003;

Goff et al., 2006). Submarine sediment slides (Bondevick et al., 2005), onshore landslides (Pararas-Carayanni, 2003), and subsea volcanic eruptions (Sacchi et al.,

2005; Satake, 2007) can also produce large and widespread tsunamis.

Microfossils incorporated into coastal sediments provide an independent test of earthquake related RSL change and tsunami deposition (Fig. 2.1). Diatoms—long recognized as valuable RSL indicators—have been applied to earthquake and tsunami studies around the world for over two decades (e.g., Hemphill-Haley, 1995a, 1995b;

Sawai et al., 2008; Briggs et al., 2014; Dura et al., 2014) (Fig. 2.2). Early earthquake studies attempted to quantify sudden coseismic RSL change based on plant macrofossils in buried soils, but the broad elevational range of macrofossil species introduced large errors (>0.5 m) to estimates (Atwater, 1987, 1992b; Clague and Bobrowsky, 1994). Early, semi-quantitative diatom-based approaches (Darienzo et al., 1994, Hemphill-Haley,

1995a, Nelson et al., 1996a, and Shennan et al., 1996) offered higher levels of precision.

In the last ten years, quantitative diatom-based reconstructions (e.g., transfer functions) have used the ever-growing knowledge of the modern distribution of diatoms in tidal marshes to estimate coseismic RSL change, often with decimeter scale errors (Zong et

23 CHAPTER 2 al., 2003; Sawai et al., 2004a,b; Shennan and Hamilton, 2006; Nelson et al., 2008;

Watcham et al., 2013).

The application of diatoms in tsunami studies focuses on the origin of anomalous sand beds that are deposited in low-energy coastal environments such as coastal marshes

(Hemphill-Haley, 1995a, 1996; Benson et al., 1997; Nanayama et al. 2007), swales

(Dawson and Stewart, 2007; Jankaew et al., 2008); lagoons (Nichol et al., 2007; Sawai et al., 2009; Wilson et al., 2014) and lakes (Hutchinson et al., 1997; Grauert et al., 2001;

Kelsey et al., 2005). Allochthonous marine assemblages within terrestrial settings may be indicative of short-lived, abrupt marine incursions from tsunamis (Goff et al., 2012).

In addition, the quality of preservation and distribution of diatoms throughout the deposit can be used to infer high-energy transport (Hemphill-Haley, 1996; Sawai et al.,

2002). Early applications of diatoms to tsunami studies focused on prehistoric and historical events (e.g. Minoura and Nakaya, 1991; Dawson et al., 1996a,b; Dawson and

Smith, 2000; Dominey-Howes et al., 2000; Chagué-Goff et al., 2002; Smith et al., 2004;

Bondevik et al., 2005; Kortekaas and Dawson, 2007). Several recent tsunamis, however, have provided an opportunity to characterize the diatom composition of modern deposits before taphonomic alteration, including the 1998 Papua New Guinea (Dawson,

2007), 2004 Indian Ocean (Sawai et al., 2009), 2009 South Pacific (Chague-Goff et al.,

2011), 2010 Maule Chile (Horton et al., 2011; Garrett et al., 2013) and 2011 Tohoku-Oki

(Szczuciński et al., 2012; Sawai et al., 2012) tsunamis.

In this review, we illustrate varying approaches of diatom research that have produced observations, analyses and interpretations regarding earthquakes and tsunamis. We discuss examples that utilized diatoms to estimate earthquake related RSL changes, in

24 CHAPTER 2 particular recent statistically-based approaches. We explore the application of diatoms to paleotsunami and historical tsunami studies and discuss the advancements and the challenges in using diatoms as tsunami indicators. Finally, we discuss the knowledge gaps and limitations of diatom research, and make recommendations for future earthquake and tsunami studies.

2.2 DIATOMS AND COASTAL ENVIRONMENTS

Diatoms are photosynthetic, unicellular algae ranging in size from 5 μm to 200 μm that inhabit freshwater, brackish, and marine environments (Jones, 2007). They are generally the dominant microphyte in estuarine and marine littoral environments and their siliceous valves are resistant to taphonomic degradation and oxidation (Admiraal,

1984; Palmer and Abbott, 1986; Cooper et al., 2010). As a result, small sample sizes (i.e. core studies) are applicable for study because they contain statistically significant diatom populations (Birks, 1995).

Diatoms have been described and classified for over 200 years based on their different shapes, sizes, and the intricate morphological characteristics of their siliceous valves

(Round et al., 1990). Currently, diatom taxa are divided into three main classes: the

Coscinodiscophyceae (centric taxa); Fragilariophyceae (araphid pennate taxa); and

Bacillariophyceae (raphid pennate taxa). Commonly used literature for the identification of diatoms to species level includes van der Werff and Huls (1958-1974), Patrick and

Reimer (1966, 1975), Krammer and Lange-Bertalot (1986, 1988, 1991a,b), Hartley et al.

(1996), Lange-Bertalot (2000) and Horst and Lange-Bertalot (2000-2013).

Diatoms are a valuable tool in reconstructing paleoenvironmental changes because of their preferences to a number of environmental factors including salinity, tidal exposure,

25 CHAPTER 2 substrate, vegetation, pH, nutrient supply and temperature. Comprehensive ecological classifications such as those by Hustedt (1937, 1939, 1953, 1957), Lowe (1974), Patrick and Williams (1990), Denys (1991-1992), Juggins (1992), Vos and de Wolf (1988; 1993), and Lange-Bertalot (2000) classified diatom taxa based on how they respond to these environmental factors.

Diatoms’ preferences for salinity and their life form are particularly valuable for earthquake and tsunami studies (Fig. 2.3). Changes in salinity across the intertidal zone produces vertically zoned diatom assemblages with respect to the tidal frame (Hemphill-

Haley 1995b; Sherrod et al., 1999; Horton and Sawai, 2010) that are prerequisite for reconstructing RSL (e.g., Zong and Tooley, 1996). Polyhalobous and mesohalobous diatom taxa represent the marine and brackish conditions found in tidal flats and lower tidal marshes and mangroves. Oligohalobous halophile and oligohalobous taxa become dominant through the transition from tidal marsh/mangrove to freshwater environments, and halophobous taxa characterize the most landward freshwater communities above the highest tides (e.g. Nelson and Kashima, 1993; Sherrod, 1999;

Zong and Horton, 1998, 1999; Sawai, 2001; Patterson et al., 2005; Szkornik et al., 2006;

Horton et al., 2007; Woodroffe and Long, 2010).

Figure 2.3 illustrates diatom assemblages from salt marshes in eastern Hokkaido, northern Japan (Sawai et al., 2004b) and southern Washington State, USA (Hemphill-

Haley, 1995a,b). The assemblages show a clear transition from subtidal, open-water marine diatoms (e.g., Thalassiosira lacustris and Odontella aurita), to typical marine tidal flat taxa (e.g., Achnanthes brevipes and Tabularia fasciculata). In the low marsh, where a mixed diatom community is found, marine–brackish diatoms (e.g.

26 CHAPTER 2 Planothidium delicatulum and Tryblionella granulata) dominate, followed by freshwater and salt-tolerant freshwater taxa in the high marsh (e.g., Caloneis bacillum and Cosmioneis pusilla) and upland (e.g., Eunotia bilunaris and Aulacoseira crassipunctata).

Diatoms are also distributed across the intertidal zone based on their affinity for certain substrates. In tsunami studies, the life-form of diatoms is often employed to support the allochthonous nature of inferred tsunami deposits. Hustedt (1958), Vos (1986), Vos et al.

(1988), and Vos and de Wolf (1993) define diatom life-forms based on the substrate—or lack thereof—that particular diatom taxa commonly live on. Benthic diatoms are grouped into epipsammic taxa that live attached to sand grains; epipelic taxa that live on or just below the surface of wet muddy sediment; epiphytic taxa that are attached to larger plants or other surfaces; and aerophile taxa that are able to survive subaerial, temporarily dry conditions. Planktonic diatoms float freely in the water column and do not live attached to any substrate; tychoplanktonic diatoms include an array of species that live in the benthos, but are commonly found in the plankton. Based on local conditions, epipsammic, epipelic, planktonic, and tychoplanktonic diatoms may comprise tidal flat populations, whereas epiphytic and epipelic forms are more common on tidal marshes/mangroves, and aerophilous forms are most common within the landward communities above the highest tides (Sherrod, 1999).

2.3 APPLICATION OF DIATOMS TO PALEOSEISMIC STUDIES

2.3.1 Earthquake deformation cycle

On subduction zone coastlines, the earthquake deformation cycle describes the characteristic and repeated RSL changes associated with great earthquakes (Nelson et

27 CHAPTER 2 al., 1996b). The nature of the interseismic and coseismic motion of the coastline is determined by its proximity to the trench, the geometry of the subduction zone and where ruptures stop along strike (Plafker, 1965; Plafker and Savage, 1970; Briggs et al.,

2014). For example, some subduction zone coastlines (e.g. Cascadia) lie within a zone that gradually uplifts in between earthquakes and abruptly subsides during great earthquakes (Nelson et al., 1996b). In coastal sedimentary sequences, the interseismic period (lasting centuries to millennia) is represented by a gradual regression from clastic to organic-rich sediments as RSL falls and marine influence decreases. Abrupt coseismic land subsidence creates a transgression as organic-rich units are lowered into the intertidal zone and overlain with a sharp contact by clastic sediment, reflecting a RSL rise and an increase in marine influence (Fig. 2.1).

Conversely, some coastal locations bordering subduction zones (e.g. eastern Aleutian megathrust) lie within a zone that gradually subsides in between earthquakes and abruptly uplifts during great earthquakes (Shennan et al., 2009). Here, during the interseismic period, coastal sedimentary sequences display a gradual transgression from organic-rich to clastic sediments as RSL rises and marine influence increases. Abrupt coseismic uplift creates a sudden regression as clastic sediments are uplifted higher in the tidal frame and organic-rich sediments form, reflecting a RSL fall and a decrease in marine influence. In addition, coastal sedimentary sequences can record both coseismic uplift and subsidence at one location, depending on the amount and distribution of slip upon the megathrust surface (Briggs et al., 2014; Ely et al., 2014). The possibility of a variable uplift-subsidence history illustrates how the paleoseismic record at individual sites should not be assumed to record exclusively coseismic uplift or subsidence.

28 CHAPTER 2 Because non-seismic coastal processes can also produce changes in lithology similar to those created by great subduction zone earthquakes, criteria were developed to help support an earthquake deformation cycle origin (Darienzo et al., 1994; Nelson et al.,

1996b; Shennan et al., 1996; Dura et al., 2014). The key criteria are the suddenness of

RSL change; the lateral extent of the lithologic contacts; the amount of RSL change; the synchroneity of RSL change among regional sites; and the coincidence of tsunami deposits with sudden changes in lithology. Although many paleoseismic studies have relied on stratigraphic investigations to support an earthquake origin of lithologic contacts (Atwater, 1987; Dura et al., 2011), diatoms have been particularly useful in providing evidence to support these inferences, because of their utility in identifying environmental changes caused by RSL change throughout the earthquake deformation cycle (Atwater and Hemphill-Haley, 1997).

2.3.2 Diatoms and relative sea-level changes related to the earthquake deformation cycle

Reconstructing and quantifying RSL change is based on the relationship between diatom assemblages and elevation with respect to the tidal frame (e.g., Zong and Horton, 1998).

Over time, diatoms become incorporated in coastal sediments, resulting in buried assemblages that represent an environmental history that can span thousands of years.

Sudden changes in diatom assemblages across transgressive or regressive contacts can signal an abrupt change in depositional environment from earthquake-related RSL rise

(land-level subsidence) or fall (land-level uplift), and can be used to estimate the amount of coastal deformation (Table 2.1). Much of the early research using diatoms to reconstruct RSL changes related to the earthquake deformation cycle focused on

Cascadia. Darienzo and Peterson (1990) and Darienzo et al. (1994) employed qualitative 29 CHAPTER 2 diatom analyses to confirm inferences of sudden and widespread coseismic subsidence based on distinctive lithologic contacts. To approximate coseismic land-level change,

Darienzo et al. (1994) estimated the elevation represented by diatom assemblages within the pre-earthquake organic-rich units, and overlying post-subsidence clastic sediments with respect to modern tide levels. But, the limited modern intertidal diatom data available for Cascadia (e.g., Riznyk, 1973; Amspoker and McIntire, 1978; Whiting and

McIntire, 1985) introduced large (> 1 m) errors to land-level change reconstructions.

Quantitative methods employed the relationship between modern diatom distributions and environmental variables such as elevation and salinity (e.g. Nelson and Kashima,

1993; Hemphill-Haley 1995b; Shennan et al., 1996; Sherrod et al., 1999) in order to reconstruct RSL changes in fossil sections in Cascadia (Hemphill-Haley, 1995a; Nelson et al., 1996a; Shennan et al., 1996; Sherrod et al., 2000; Kelsey et al., 2002; Witter et al.,

2003). In a study of modern diatom distribution in intertidal marshes and tidal flats in southern Washington State, USA, Hemphill-Haley (1995b) used factor analysis to identify the prevalent occurrences in elevational zones relative to tidal level of taxa also commonly found in the subsurface (fossil) record (Fig. 2.3). The result was a list of taxa with dominant occurrences in marsh zones (either high marsh, low marsh, or both high and low marshes) or three subenvironments of the lower intertidal to shallow subtidal zone (mud flats, sand flats, and Zostera (eelgrass) beds; Hemphill-Haley 1995a). The modern distributions of the same taxa found in subsurface stratigraphy were used to reconstruct the abrupt changes in environment indicated by lithologic boundaries spanning an inferred coseismic subsidence event that occurred ~300 years ago (Fig. 2.4).

Distinct changes in diatom assemblages across the peat-mud contact confirmed significant (0.8-3.0 m), widespread, and lasting change from an upland environment to a

30 CHAPTER 2 tidal-flat or low marsh environment, consistent with rapid coseismic subsidence.

Shennan et al. (1996) expanded the quantitative approach by employing statistical methods that directly compared modern and fossil diatom assemblages and improved the precision of quantitative estimates of earthquake related land-level change to ± ~0.5 m.

The development of larger modern training sets that spanned a greater range of intertidal environments allowed for the application of transfer function techniques in

RSL reconstructions (Guilbault et al., 1995; 1996; Hughes et al., 2002; Hawkes et al.,

2011; Engelhart et al., 2013). Transfer functions use multivariate statistical techniques to formalize the relationship between the relative abundance of microfossil species and an environmental variable of interest (in the case of paleoseismic studies, that variable is elevation) (e.g., Horton and Edwards, 2006). The transfer function is then applied to microfossil assemblages in sediment cores in order to calculate changes in paleo- elevation (Horton et al., 1999; Kemp et al., 2009). The performance of transfer functions can be measured using multiple techniques to assess the effect of sample design, the goodness-of-fit between elevation and diatom assemblages, and the statistical significance of each reconstruction (Telford and Birks 2011a, 2011b). The first application of a transfer function to produce a record of earthquake-related RSL change was using the foraminiferal microfossil group in Cascadia (Guilbault et al., 1995; 1996).

The technique has since been expanded to include diatoms (Sherrod, 2001) and other geographical locations (Shennan and Hamilton, 2006) providing a continuous record of

RSL changes throughout the earthquake deformation cycle (Table 2.1, Fig. 2.2; Sawai et al., 2004b).

31 CHAPTER 2 Zong et al. (2003) used detailed modern and fossil diatom analyses from a tidal marsh along the eastern Aleutian megathrust to produce a precise record (with errors of a few decimeters) of RSL changes during parts of six earthquake deformation cycles. In addition to recording abrupt subsidence during great earthquakes, the diatom data also identified small amounts of subsidence in the years prior to some of the events, interpreted by Zong et al., 2003 as a preseismic signal. If preseismic deformation does indeed occur, the implication is that warning signs that are detectable for several years prior to a great earthquake (Bourgeois, 2006). To explore the possibility that the pre- seismic signal was a result of downward mixing of diatom assemblages, either through biological or physical processes, Hamilton et al. (2005) transplanted a block of marsh peat to a lower elevation in the intertidal zone where it would be buried by tidal mud.

The results showed that mixing of diatoms did occur, but only in the top ~ 1 cm of peat, whereas the pre-seismic signal observed by Zong et al. (2003) and Hamilton and

Shennan (2005) occurred over 2-5 cm. In addition, Shennan and Hamilton (2006) argued that the pre-seismic signal is not a result of diatoms filtering down from overlying mud because dominant species that reflect subtle pre-seismic subsidence, such as

Nitzschia obtusa, Navicula begeri, Navicula brockmanii and Pinnularia lagerstedtii, do not occur in the overlying mud (Fig. 2.5).

Coastal stratigraphy from eastern Hokkaido, Japan contains evidence for cycles of coastal deformation (Sawai et al., 2002, 2004a, b) associated with multi-segment earthquakes originating from the Kuril trench (Nanayama et al., 2003). Sawai et al.

(2004a) used a detailed modern diatom study (Sawai et al., 2001b, 2004b) to create a diatom-based transfer function documenting pre-seismic, coseismic and post-seismic land-level change associated with a great 17th century earthquake (Fig. 2.6). Fossil

32 CHAPTER 2 diatom assemblages suggest that tidal flats gradually changed into freshwater upland in decades after the 17th century tsunami as a result of up to 1.5 m of post-seismic deformation. The tsunami and the ensuing uplift exceeded any in the region’s 200 years of written history, and both resulted from a plate-boundary earthquake of unusually large size along the Kuril subduction zone (Atwater et al., 2004, Sawai et al., 2004a).

2.4 APPLICATION OF DIATOMS TO TSUNAMI STUDIES

2.4.1 Paleotsunamis

Well-preserved sequences of tsunami deposits are a geological record of their source events (e.g. earthquakes, landslides, and volcanic eruptions) and can be used to estimate event recurrence intervals (Nanayama et al., 2003; Cisternas et al., 2005; Jankaew et al.,

2008; Sawai et al, 2012). Sedimentary sequences tell of repeated Holocene tsunamis

(Table 2.1, Fig. 2.2) from the North Sea (Bondevik et al., 2005), Greece (Dominey-Howes et al., 2000), New Zealand (Goff et al., 2001; Cochran et al., 2006, 2007; Nichol et al.,

2007), Kamchatka (Bourgeois et al., 2006; Martin et al., 2008), the southern Kuril

Trench (Sawai, 2002; Nanayama et al. 2003; Iliev et al., 2005), the Japan Trench

(Minoura and Nakaya, 1991; Minoura et al., 1994; Sawai et al., 2008, 2012), the Aleutian

Megathrust (Briggs et al., 2014; Shennan et al., 2014b), the Cascadia subduction zone

(Clague et al., 1999; Kelsey et al., 2005; Williams et al., 2005; Nelson et al., 2008), the

Chilean subduction zone (Cisternas et al., 2005; Horton et al., 2011; Garrett et al., 2013;

Dura et al., 2014), and the Sunda megathrust (Jankaew et al., 2008; Monecke et al.,

2008; Sawai et al., 2009).

Sand and silt beds deposited by tsunamis in low-energy coastal environments often display characteristics such as significant (local and regional) lateral extent, uniform

33 CHAPTER 2 thickness, sharp (1-3 mm) lower and upper contacts, landward thinning, and normal grading (upward fining) of grain size (Dawson and Stewart, 2007; Szczuciński et al.,

2012). Allochthonous marine and brackish diatoms within tsunami deposits, especially offshore planktonic taxa, can support a seaward source of the sediment (e.g., Goff et al.,

2012; Szczuciński et al., 2012; Dura et al., 2014). For example, diatom analysis revealed three sandy deposits containing marine and brackish diatoms at Suijin-numa, a coastal lake midway along the subduction zone marked by the Japan Trench (Sawai et al.,

2008). The marine and brackish diatoms (Diploneis smithii, Delphineis surirella) within the sand beds contrasted against the freshwater assemblages (Aulacoseira granulata, A. crassipunctata, Eunotia spp.) in the mud under-and-overlying the beds, supporting a marine origin of the sand. Age models suggest the middle sand bed correlates with the

Jogan tsunami in AD 869 (Yoshida, 1906) that reportedly devastated at least 100 km of coast approximately centered on Sendai and is the presumed predecessor to the devastating 2011 Tohoku-oki earthquake in Japan. Along the Kuril Trench in northern

Japan, Nanayama et al. (2007) used diatoms to identify nine sandy tsunami deposits intercalated with peat. The diatom assemblage within the peats contained freshwater species (e.g., Eunotia spp., Pinnularia spp.), whereas sand beds were dominated by marine taxa including Delphineis surirella and Odontella aurita. Based on the record of tsunami deposits, the authors estimated a 365 – 553 year recurrence interval for large

Kuril Trench earthquakes. Later, based on over 60 radiocarbon age estimations, Sawai et al. (2009) found that the interval between outsized tsunamis on the Kuril trench ranged from 100 to 800 years, with an average recurrence interval of ~400 years.

Sand beds deposited by the Storegga Slide tsunami in the North Sea also have a distinct marine signature that differentiates them from under-and-overlying sediments (Dawson

34 CHAPTER 2 et al., 1996a; Smith et al., 2004). Dawson et al. (1996a) found not only an anomalous marine diatom assemblage within Storegga Slide tsunami sediments in Scotland, but also reported high occurrences of fragmented diatoms (up to 90% of elongate valves) within sand beds reflecting the nature of the rapid, turbulent marine incursion of tsunami events (Fig. 2.7). On Chirikof Island, Alaska, USA, a sequence of anomalous sand beds intercalated with peat contained a freshwater diatom assemblage very similar to under-and-overlying peat assemblages (Nelson et al., 2014, in review). In the absence of anomalous marine diatoms within sand beds, Nelson et al. (2014, in review) used the high fragmentation and lower concentration of valves in the beds, combined with geochemical, anthropological, and historical-analog data to show deposition by tsunami.

Diatom assemblages within tsunami deposits are, however, often composed of mixed assemblages because tsunamis erode, transport, and deposit marine, brackish, and freshwater sediments (with associated taxa) as they inundate coastal and inland areas

(e.g., Dawson et al., 1996a,b; Dawson and Smith, 2000; Smith et al., 2004; Bondevik et al., 2005; Sawai et al., 2008). Therefore, the identification of paleotsunamis can be a challenge. Evidence of coseismic land-level change coincident with anomalous sand beds provides a definitive earthquake source for a tsunami. Hemphill-Haley (1995a) found a well-preserved diatom assemblage dominated by epipsammic tidal flat diatoms in a thin layer of silt and fine sand overlying a freshwater upland soil in Washington State USA.

The upland soil was displaced into the intertidal zone by coseismic subsidence during the

AD 1700 earthquake on the Cascadia subduction zone (Fig. 2.4). The coarser-grained tsunami deposit capping the soil contrasted in both grain size and diatom content to the dominantly muddy deposits between the buried soil and the modern marsh surface.

Hemphill-Haley (1995a) also used the distribution of diatoms to show that the

35 CHAPTER 2 inundation area of the AD 1700 tsunami was larger than the distribution inferred from the coarser-grained deposit visible in outcrop, tracing epipsammic tidal flat diatoms about 1 km farther upstream from the landward extent mapped in cores and stream channel outcrops. At a coastal lowland in central Chile, Dura et al. (2014) used diatoms to identify six Holocene paleotsunami deposits coincident with coseismic uplift.

Although the sand beds contained mixed diatom assemblages similar to under-and- overlying sediment, anomalous marine planktonic diatoms within the beds, along with diatom evidence of coseismic uplift concurrent with sand bed deposition, provided evidence for repeated tsunamis. Cisternas et al. (2005) employed a similar method in south-central Chile to identify historical and prehistoric tsunami deposits coincident with coseismic subsidence.

2.4.2 Contemporary tsunamis

Microfossil analysis of recent tsunami deposits provides a basis against which paleo- deposits can be compared. Dawson (2007) explored the condition of diatom valves in the

1998 Papua New Guinea tsunami deposit and found that linear, sigmoid and clavate diatoms were more readily fragmented due to their relatively fragile valve structure. In contrast, high abundances of taphonomically unaltered (i.e., pristine) diatom valves were found in tsunami deposits from the 2004 Indian Ocean tsunami in Thailand (Sawai et al., 2009), and the 2010 Maule Chile tsunami (Horton et al., 2011). Anomalously low breakage of diatoms in tsunami deposits has similarly been reported in paleotsunamis from the Pacific coast of Washington State and Puget Sound, USA (Hemphill-Haley,

1996), and it is considered to result from rapid sedimentation and burial by tsunami.

36 CHAPTER 2 In Thailand, the 2004 Indian Ocean tsunami deposited a sand bed that graded into a thin mud layer and contained a mixed diatom assemblage indicating the incorporation of sediment from numerous sources (Fig. 2.8; Sawai et al., 2009). The upward fining grain size in the sand bed was reflected by similar grading of diatom valves. The lower section of the deposit was dominated by larger epipsammic marine diatoms, whereas the middle section contained abundant marine planktonic species, and the mud cap was dominated by a mixture of smaller freshwater, brackish, and marine species, reflecting the variable flow speed of the tsunami. Diatom analysis by Chagué-Goff et al. (2011) following the

2009 South Pacific tsunami deposit in Samoa and Szczuciński et al. (2012) following the

2011 Tohoku-Oki tsunami on the Sendai plain also revealed grading of diatom valves.

However, Szczuciński et al. (2012) found no marine and few brackish diatoms in sand beds deposited on the Sendai plain by the Tohoku-Oki tsunami. Grain-size analyses of the Tohoku-Oki sand beds showed that within 1 km of the coast, beds were derived mainly from the beach and coastal dunes, and because coastal sediments contained few diatoms, the assemblage was composed of a low concentration of mostly freshwater and few brackish species sourced from the coastal plain (Szczuciński et al., 2012). Further inland the diatom concentration increased, and the assemblage and condition of valves in the deposit (% fragmentation of valves) were very similar to the underlying soil and nearby freshwater canal, suggesting the deposit was locally sourced and not transported from the coast. The similar degree of fragmentation in the underlying soil and overlying tsunami deposit shows that fragmentation is highly dependent on source material, and may sometimes be caused by other processes including natural chemical dissolution, compaction, and sample treatment (Kamatani, 1982; Cooper et al., 2010), thus the metric should be applied with caution on paleotsunami deposits.

37 CHAPTER 2 2.5 CONCERNS AND RECOMMENDATIONS

Microfossils continue to improve our understanding of earthquakes and tsunamis, their impact on coastal evolution, and how they are recorded within coastal sedimentary sequences. When using diatoms for reconstructing earthquake related RSL change, several matters must be kept in mind. The relation of diatoms to tidal elevation is not well defined in many locations due to a scarcity of modern diatom data (e.g., Dura et al.,

2014). In such cases, general ecological classifications, many of which were compiled on temperate European coasts (Hartley et al., 1986; Denys 1991-1992), are often used.

However, they may not always be applicable to fossil studies from other coastlines. Even exhaustive ecological classifications of diatoms involve uncertainties, because diatom distributions are controlled by multiple environmental factors. For example, diatom species are not only closely associated with tidal elevation and salinity, but are also sensitive to substrate, vegetation, pH, nutrient supply, and temperature (Lowe 1974;

Patrick and Williams, 1990; Denys, 1991-1992; Juggins, 1992; Vos and de Wolf, 1988;

1993; Lange-Bertalot, 2000). Subtle differences in marsh vegetation and substrate at two adjacent marshes can result in changes in the distribution of diatoms across intertidal zones (Sawai et al., 2004b). Thus, more modern diatom studies are needed to improve qualitative or quantitative interpretations of RSL change based on fossil diatom assemblages.

However, even large modern training sets that include hundreds of samples taken across multiple local marshes can sometimes fail to include modern analogs for fossil diatom assemblages (Watcham et al., 2013). This is known as the non-modern-analogue situation (Birks, 1995). This situation occurs when local environmental conditions have changed significantly over time. Coastal sites have experienced changes in hydrology,

38 CHAPTER 2 sedimentary regime and morphology throughout the Holocene, due to changes in sea level (tectonic and eustatic) and anthropogenic forces. Consequently, the diatom assemblages from a modern environment may not match those in a fossil one in the same locality (Horton and Edwards, 2005; Wilson and Lamb, 2012). If this happens, it is not appropriate to apply quantitative reconstruction techniques (e.g., transfer function) using only local modern diatom assemblages (Watcham et al., 2013). Regional modern training sets compiled from a large range of marshes (e.g. marshes with a variety of vegetation zones, substrates, and elevation gradients) can account for such variation in the distribution of diatoms, providing analogs for diatom assemblages found in fossil cores (Zong et al., 2003; Watcham et al., 2013; Shennan et al., 2014a).

Other challenges in using diatoms to reconstruct earthquake related RSL change include variable diatom production and preservation, and the difficulty of differentiating autochthonous diatoms (in situ) and allochthonous diatoms (transported) in sediment samples. In the Copper River Delta, Shennan et al. (2014a) found limited diatom abundance in surface samples taken from unvegetated tidal flat silts due to low diatom production caused by very high sediment concentration in the delta. Silts in fossil sequences also contained few diatoms, which complicated quantitative reconstructions of RSL (Shennan et al., 2014a). Allochthonous diatoms pose the opposite problem, as certain taxa are transported throughout the marsh and are subsequently included in fossil assemblages that do not accurately represent the paleoenvironment. This is especially common in coastal environments, where daily tidal currents, or in colder climates, sediment laden sea-ice rafts, transport diatom valves from one marsh elevation zone to another (Hemphill-Haley et al., 1995; Hamilton et al., 2005). Several ways have been suggested to distinguish allochthonous diatoms from fossil assemblages (e.g. Sawai,

39 CHAPTER 2 2001). One approach is to regard planktonic diatoms that can be transported by tidal currents as allochthonous components in fossil assemblages, while benthic (e.g. epiphytic, epipelic and epipsammic types) are considered as autochthonous (Vos and de

Wolf, 1993). Another solution to the autochthonous/allochthonous problem is to exclude potential allochthonous species from the data interpretation. For example, Hemphill-

Haley, 1995a excluded the marine planktonic diatom Paralia sulcata from paleoecological interpretations, because its robust valves and long-chained structure allows its valves to be easily floated and transported by tidal currents and deposited far inland in tidal estuaries (Hemphill-Haley, 1995a) and into isolation basins (Zong, 1997).

Another example is Cocconeis scutellum, an epiphytic species living on macrophytes.

After their death, non-attached C. scutellum rapheless valves become separated from the macrophytes and can then be transported over long distances selectively by tidal currents (Sawai, 2004b). Consequently, rapheless valves can be found in sediment across the entire tidal environment, although its habitat is limited to the macrophytes zone

(Sawai, 2001b).

The geographical limitations of our understanding of the relation of diatoms to tidal elevation also complicate the interpretation of tsunami deposits. Identifying an anomalous, allochthonous assemblage is difficult if regional diatom ecology is limited.

Preservation is an issue in tsunami deposits as well, due to both source material containing few diatoms (e.g. Szczuciński et al., 2012) and chemical dissolution of valves after they are incorporated into a deposit (Jankaew et al., 2008). Sawai et al. (2009) found excellent preservation of diatoms in the 2004 Indian Ocean tsunami deposit in

Thailand, but three paleotsunami deposits examined at the same site by Jankeaw et al.

(2008) contained no fossil diatoms. The lack of diatoms in the Thailand fossil sections

40 CHAPTER 2 may be the result of chemical dissolution, which Katamani (1982) showed is positively correlated to temperature (Kamatani, 1982).

Distinguishing between sand beds deposited during the largest storms and by tsunamis is a major concern, although general discriminating criteria have been proposed

(Nanyama et al., 2000; Witter et al., 2001; Goff et al., 2004; Tuttle et al., 2004; Morton et al., 2007; Kortekaas and Dawson, 2007; Switzer and Jones, 2008). Like tsunamis, the identification of storm-surge deposits is based on anomalous sand beds in an environment where they are unusual, such as lakes and marshes (e.g. Liu and Fearn,

1993; van de Plassche et al., 2004; Donnelly and Woodruff, 2007; Horton et al., 2008), and the diatom species present in the storm layer reveal some information about the source of the sediment. Parsons (1998) reported a multi-source origin for sediments deposited by Hurricane Andrew when it made landfall in Louisiana in 1992. Diatom assemblages within the storm-surge deposit in a salt marsh pond were diverse and consisted of a mixture of diatoms from marine, brackish, and freshwater settings and were easily distinguished from underlying autochthonous salt-marsh taxa. However, as highlighted by Hemphill-Haley et al. (1995a), in most cases it would be impossible to differentiate a storm-surge deposit from a tsunami deposit based on diatoms alone. As discussed in Section 4.1, the strongest evidence for sediment deposition by tsunami that diatoms can provide is evidence for land-level change coincident with sand bed deposition. The identification of tsunami deposits, especially in the absence of land-level change, requires a multi-proxy approach combining of microfossil, geochemical, anthropological, and historical-analog data (Goff et al., 2012).

41 CHAPTER 2 Finally, the long-term relative sea-level histories of coastlines must be considered when interpreting diatom evidence of past earthquakes and tsunamis. Along coastlines that record a mid-Holocene highstand (Mitrovia and Milne, 2002; Peltier, 2004; Milne et al.,

2005) stratigraphic evidence of earthquakes and tsunamis is often found during a limited period when the eustatic contribution to sea level was dominant and net sea-level rise created low-energy depositional environments that provided the accommodation space necessary for preservation (e.g., Dura et al., 2011; Brill et al., 2013). Thus, fossil diatom assemblages will reflect distinctly different paleoenvironments than are found in the present day. In addition, higher past relative sea levels must be considered when inferring the magnitude of tsunamis based on the presence of sand beds, as low tsunamis that may not have a significant impact on the coast today may have been able to inundate lowlands in the past.

2.6 CONCLUSIONS

Diatoms help identify and characterize evidence of earthquakes and tsunamis contained in coastal sedimentary sequences. The utility of diatoms as sea-level indicators in low- energy coastal sedimentary deposits is underpinned by the concept that characteristic assemblages have strong associations with salinity, tidal elevation, and substrate and that this modern, observable relationship has existed without change during the time represented by coastal sedimentary sequences. We have provided examples from Chile, the Indian Ocean, Japan, New Zealand, the North Sea, the South Pacific, and the United

States (Alaska and the Pacific Northwest) that illustrate how diatoms from coastal environments have been used to reconstruct RSL change associated with coseismic subsidence and uplift, and tsunami inundation along subduction zone coastlines. These examples illustrate how capturing centennial to millennial event records through

42 CHAPTER 2 stratigraphic and diatom analysis can provide important information regarding the magnitude and frequency of pre-historic earthquakes and tsunamis.

We have also highlighted the variability of diatom evidence for earthquake related RSL change and tsunami inundation and discussed the importance of understanding the geological and geomorphological context of sampling locations, and the processes operating within them when making paleoecological interpretations. Many of the disadvantages to diatom-based earthquake and tsunami reconstructions can be addressed by continuing to explore the modern associations of diatoms to environmental variables, thereby improving the accuracy of paleoecological interpretations. In addition, the continued study of the response of modern diatoms to changes in RSL and of the diatom composition and taphonomic character of modern tsunami deposits is critical to improving paleoearthquake and tsunami reconstructions.

2.7 ACKNOWLEDGEMENTS

This work was supported by funding from National Science Foundation (EAR-1144537,

1419824, 1357722) awarded to BPH. This paper is a contribution to IGCP project 588

‘Preparing for coastal change’. Some symbols in figures were courtesy of the Integration and Application Network (ian.umces.edu/symbols/), University of Maryland Center for

Environmental Science.

43 CHAPTER 2

Figure 2.1: Schematic drawing of coseismic uplift (a) and subsidence (b) and accompanying tsunami inundation. (1) When the strain is released during an earthquake, the coast suddenly subsides (a) or uplifts (b); (2) If the earthquake has produced a tsunami, it will impact the coast minutes to hours after the earthquake; (3) During the interseismic period, strain accumulates between the upper plate and subducting plate causing gradual coastal subsidence (a) or uplift (b).

44 CHAPTER 2

45 CHAPTER 2 Figure 2.2 (previous page): Summary of paleoseismic studies applying diatoms referenced in this paper. In: Indian Ocean; Ja: Japan Trench; Ch: Chilean Subduction zone; Al: Aleutian megathrust; Ca: Cascadia subduction zone; NZ: New Zealand; NS: North Sea; SP: South Pacific; PNG: Papua New Guinea. Site labels correspond to study labels in Table 2.1.

Figure 2.3: (a) Distribution of modern diatoms in salt marshes of eastern Hokkaido, northern Japan and Washington State, USA. (Modified from Sawai et al., 2004 and Hemphill-Haley, 1995a).

46 CHAPTER 2

Figure 2.4: Diatoms evaluated relative to modern intertidal zones and subsurface stratigraphy at the Niawiakum River in Washington State, USA. (a) Position of the study area relative to the Cascadia subduction zone (barbed line), the boundary between North America and Juan de Fuca/Gorda plates that extends from the northern end of the San Andreas Fault (SAF) to the southern end of the Queen Charlotte Fault (QCF); (b) Locations of vertical sections sampled for diatoms in the Niawiakum River Valley. Sites 1-4 are cutbank outcrops exposed during low tide; (c) Location of the Niawiakum River Valley, on the eastern side of Willapa Bay; (d) Changes in diatom assemblages within and above a former upland soil buried by coseismic subsidence during a Cascadia subduction zone earthquake in AD 1700. Changes in diatom assemblages are consistent with an abrupt change from upland forest to tidal flat or low marsh. (Modified from Hemphill-Haley, 1995a). 47 CHAPTER 2

Figure 2.5: Diatom analyses during past earthquake cycles suggesting pre-seismic movement. (a) Location of south-central Alaska, USA; (b) area subsided in the AD 1964 earthquake (Plafker, 1969); (c) RSL changes reconstructed using a diatom-based transfer function; two short periods of pre-seismic submergence immediately prior to the substantial co-seismic subsidence were recognized at the top of the peat unit and are highlighted. (Modified from Shennan, I., Hamilton, S., 2006).

48 CHAPTER 2

Figure 2.6: Land-level reconstructions in Hokkaido, northern Japan. (a) The solid line with triangles shows the seaward edge of the subduction zone. The volcanoes responsible for tephra layers in c and d are shown and rupture areas of instrumentally recorded earthquakes on the plate boundary off eastern Hokkaido are outlined; (b) Index map; (c) Stratigraphic cross-section; (d) Photograph of stratigraphy showing the seventeenth-century large earthquake. Example of change from tidal-flat mud to lowland- forest peat, punctuated by a tsunami deposit and by volcanic ash layers. (e) Diatom diagram showing the schematic stratigraphy, changes in diatom assemblages, and the results of diatom-based transfer functions for land-level change. Error bars for height estimates span two standard deviations. (Modified from Sawai et al., 2004a).

49 CHAPTER 2

Figure 2.7: Diatom analyses of the Storegga Slide tsunami showing fragmentation of diatom valves within the high-energy deposit. (a) Location of the Storegga Slides and sites where evidence for the Holocene Storegga Slide tsunami has been found; (b) Sites in the United Kingdom where evidence for the Holocene Storegga Slide tsunami may be found. Numbers correspond to sites discussed in Smith et al., 2004; (c) Diatom summary diagram from Boreholes 53 and 23, lower Wick River Valley, Caithness, Scotland. Taxa displayed as % of total valves (Modified from Smith et al., 2004).

50 CHAPTER 2

Figure 2.8: Diatom analyses of the 2004 Indian Ocean tsunami deposit showing grading of diatom valves; (a) Fault slip distribution during the 2004 Sumatra–Andaman earthquake has an outer contour at 5 m and inner contours at 10, 15, and 20 m (Chlieh et al., 2007); (b) Phra Thong Island. The island is isolated from mainland of Thailand by inlets. Light gray area is grassy beach ridge plains. Dark gray area is mangrove forests. Landforms traced from 1:50,000-scale airphotos taken in 1999 and from post-tsunami satellite images at PointAsia.com (modified from Jankaew et al., 2008); (c) Location of pit and modern samples. Satellite image is from PointAsia.com; (e) Simplified process of deposition of diatoms and sediment during tsunami; (1) Fast current. Only beach and subtidal species are incorporated with coarse sediment. Because turbulent current can keep substantial amount of sand fractions in the water column, mixed assemblage of many beach and subtidal, marine plankton are suspended. Freshwater specimens may be included with eroded soil fractions. (2) Current becomes slow. Fine fractions also fall onto the ground. Eroded, floated, and transported specimens are also incorporated. (3) Suspension stage (calm current) of tsunami. All floated specimens are allowed to settle down. Many freshwater species incorporated with their substrata (plant trash and eroded soil fractions). (Modified from Sawai et al., 2009).

51 CHAPTER 2 TABLE 2.1. PALEOSEISMIC STUDIES EMPLOYING DIATOMS

52 CHAPTER 2

1 H: Historical; P: Prehistoric; Qn: Quantitative; Ql: Qualitative 2 H: Historical; P: Prehistoric 3 Mx: mixed assemblage (marine/brackish/freshwater diatoms); M: marine diatoms; Fw: freshwater diatoms; M&B: marine and Brackish diatoms; Fw&B: freshwater and brackish diatoms; Ep: sand flat epipsammic diatoms; Pl: marine planktonic diatoms; G: Grading of diatom valves 4 Metric defined in individual papers. 5 Metric defined in individual papers.

53

CHAPTER 3. COASTAL EVIDENCE FOR HOLOCENE SUBDUCTION- ZONE EARTHQUAKES AND TSUNAMIS IN CENTRAL CHILE*

*Chapter accepted for publication in Quaternary Science Reviews as: Dura, T., Cisternas, C., Horton, B.P., Ely, L.L., Nelson, A.R., Wesson, R.L., Pilarczyk, J.E., Coastal evidence for Holocene subduction-zone earthquakes and tsunamis in central Chile.

ABSTRACT

The ~500-year historical record of seismicity along the central Chile coast (30-34S) is characterized by a series of ~M 8.0-8.5 earthquakes followed by low tsunamis (<4 m) occurring on the megathrust about every 80 years. One exception is the AD 1730 great earthquake (M 9.0-9.5) and high tsunami (>10 m), but the frequency of such large events is unknown. We extend the seismic history of central Chile through a study of a lowland stratigraphic sequence along the metropolitan coast north of Valparaíso (33S). At this site, higher relative sea level during the mid Holocene created a tidal marsh and the accommodation space necessary for sediment that preserves earthquake and tsunami evidence. Within this 2600-yr-long sequence, we traced six laterally continuous sand beds probably deposited by high tsunamis. Plant remains that underlie the sand beds were radiocarbon dated to 6200, 5600, 5000, 4400, 3800, and 3700 cal yr BP. Sediment properties and diatom assemblages of the sand beds—for example, anomalous marine planktonic diatoms and upward fining of silt-sized diatom valves—point to a marine sediment source and high-energy deposition. Grain-size analysis shows a strong similarity between inferred tsunami deposits and modern coastal sediment. Upward fining sequences characteristic of suspension deposition are present in five of the six sand beds. Despite the lack of significant lithologic changes between the sedimentary units under- and overlying tsunami deposits, we infer that the increase in freshwater siliceous microfossils in overlying units records coseismic uplift concurrent with the 54

CHAPTER 3 deposition of five of the sand beds. During our mid-Holocene window of evidence preservation, the mean recurrence interval of earthquakes and tsunamis is ~500 years.

Our findings imply that the frequency of historical earthquakes in central Chile is not representative of the greatest earthquakes and tsunamis that the central Chilean subduction zone has produced.

3.1 INTRODUCTION

The most densely populated portion of the Chilean coast lies along the central Chilean subduction zone, a historically active segment of the megathrust between 30 and 34S

(Fig. 3.1; Lomnitz, 2004; Metois et al., 2012). Historical accounts from central Chile describe destructive earthquakes in AD 1575, 1580, 1647, 1730, 1822, 1906, and 1985

(Lomnitz, 1970, 2004; Cisternas et al., 2012). The consistent recurrence interval (~80 years; Comte et al., 1986; Barrientos, 2007) suggests the next large earthquake will not occur until the mid-21st century. But the 2004 Andaman-Aceh (Sumatra) and 2011

Tohoku-oki (Japan) multi-segment subduction-zone earthquakes and accompanying tsunamis highlight the risk of basing hazard assessment solely on well-documented historical events (Rhodes et al., 2006; Geller, 2011; Heki, 2011; Stein and Okal, 2011).

The consistency of the historical recurrence interval in central Chile is misleading because rupture lengths and amounts of slip on the megathrust differ substantially among earthquakes (Fig. 3.1b; Moreno et al., 2007; Cisternas et al., 2012). The height of tsunamis accompanying historical earthquakes has also varied. The most recent and well-documented historical earthquakes in AD 1822, 1906 and 1985 were M 8.0-8.5 events that produced localized damage and relatively low tsunamis (<4 m). The AD 1730 earthquake, which is not as well documented because it occurred when the coastal population was sparse, caused widespread damage and produced a high tsunami on both

55 CHAPTER 3 the Chilean (Lomnitz, 1970; Udías et al., 2012) and Japanese coasts (Soloviev and Go,

1984). Based on the widespread tsunami, the 1730 earthquake was at least as large as the

Mw 8.8 2010 Maule, Chile earthquake that devastated south-central Chile. However, such an earthquake has not recurred in the last ~300 years, and, thus, is not considered to be typical of the central Chilean subduction zone (Vigny et al., 2011). In order to better assess hazards along this segment of the subduction zone, we need to know how often

1730-style earthquakes and tsunamis have occurred. Such knowledge can only be gained by extending the record of earthquakes and tsunamis in central Chile back in time.

Stratigraphic evidence of great subduction-zone earthquakes, like that discovered in the tidal marshes of Chile (Atwater et al., 1992; Bartsch-Winkler and Schmoll, 1993;

Cisternas et al., 2005; Nelson et al., 2009), Alaska (Ovenshine et al., 1976; Combellick and Reger, 1994; Shennan and Hamilton, 2006), and the U.S. Pacific Northwest

(Darienzo et al., 1994; Atwater and Hemphill-Haley, 1997; Nelson et al., 1998), can be used to develop histories of large earthquakes that extend over multiple earthquake cycles. Coseismic vertical deformation of the coastline may be recorded in coastal sedimentary sequences as a series of sudden changes in relative sea level (RSL) (Plafker and Savage, 1970; Atwater, 1987). Sudden coseismic RSL rises or falls may create a series of distinctive, sharp lithologic contacts that record earthquakes over thousands of years

(Plafker et al., 1992; Nelson et al., 1996; Witter et al., 2003). Widespread sand beds, deposited by tsunamis accompanying great earthquakes, are often found concomitant with lithologic evidence of sudden RSL change (Fig. 3.2; Sawai, 2001; Kelsey et al., 2002;

Cisternas et al., 2005; Satake and Atwater, 2007).

56 CHAPTER 3 Because of their sensitivity to environmental factors including salinity, tidal exposure, and substrate, microfossils incorporated into coastal sediments provide an independent test of earthquake-related RSL change and tsunami deposition inferred from coastal sequences (Fig. 3.2; Sherrod, 1999). Diatoms—long recognized as valuable RSL indicators—have been used to reconstruct earthquake and tsunami histories at several subduction zones (Hemphill-Haley, 1995a; Shennan et al., 1996; Sawai, 2001; Shennan and Hamilton, 2006). The utility of diatoms as RSL indicators also stems from the high preservation potential of their siliceous valves in coastal sedimentary archives (Hamilton et al., 2005; Sawai et al., 2008). Other siliceous microfossils such as phytoliths (siliceous plant remains) and chrysophyte cysts (freshwater golden algae) can be employed to similar advantage (Rebolledo et al., 2005).

Sedimentary archives of prehistoric earthquakes and tsunamis are scarce in central Chile because the combination of a semiarid climate, coseismic uplift rather than subsidence, and net Holocene emergence has resulted in limited accommodation space in which to preserve evidence (May et al., 2013). More extensive evidence has been found at wetter sites in southern and south-central Chile, although records are limited to the last ~2000 years (Atwater et al., 1992; Bartsch-Winkler and Schmoll, 1992; Cisternas et al., 2005;

Moernaut et al., 2009; Nelson et al., 2009; Garrett et al., 2013; Ely et al., 2014; Morneaut et al., 2014).

Here, we present the first stratigraphic evidence of prehistoric earthquakes and tsunamis on the coast of metropolitan central Chile (Fig. 3.3). We used sedimentary and microfossil evidence from an unusual coastal lowland at Quintero to identify six tsunami deposits between 6200 and 3600 yrs BP. We observed evidence of coseismic uplift

57 CHAPTER 3 concurrent with five of the tsunamis. The limited time window of preservation of evidence for prehistoric seismicity discovered at Quintero reflects its RSL history, whose highest RSL peaked in the mid Holocene. We show how the geomorphic setting at

Quintero created an environment favorable to the preservation of earthquake-related

RSL change and tsunami inundation about the time of the RSL highstand. The ~2600 year window provides a glimpse into the recurrence of earthquakes with high tsunamis along a coast where historical records are too short to capture the repeated occurrence of its greatest events.

3.2 TECTONIC AND COASTAL SETTING

3.2.1 Earthquake deformation cycle in central Chile

The Chilean subduction zone marks the plate boundary between the subducting Nazca plate and the South American plate, extending 3500 km from the northern coast of Chile to the Chile triple junction offshore of southern Chile (Fig. 3.1a). Along the central

Chilean subduction zone, the Nazca plate dips <10 and subducts to the northeast at 68-

80 mm/yr (Pardo et al., 2002; Métois et al., 2012). Geodetic studies show that the central Chilean subduction zone is highly coupled. Plate-boundary asperities at ~30S

(the Challenger fracture zone offshore of Tongoy) and ~33S (the Juan Fernandez Ridge offshore of Valparaíso) border the central Chilean segment and may limit rupture length

(Métois et al., 2012).

Based on their accompanying tsunamis, the four largest historical plate-boundary earthquakes in central Chile occurred in AD 1730 (M ~9.0), 1822 (M 8.0-8.5), 1906 (Mw

8.4), and 1985 (Mw 7.8) (Comte et al., 1986; Barrientos, 1995; Ruegg et al., 2009). We exclude the AD 1575, 1580, and 1647 earthquakes from this sequence because there are

58 CHAPTER 3 no accounts of accompanying tsunamis; thus, there is a possibility that they were not megathrust earthquakes (Cisternas, 2012; Cisternas et al., 2012; Udías et al., 2012). The three most recent ruptures (AD 1822, 1906, and 1985) may have been limited in their southern extent by the Juan Fernandez Ridge asperity (Fig. 3.1b), but based on its widespread effects, the AD 1730 almost certainly extended south through the asperity.

The AD 1730 earthquake affected more than 800 km of the Chilean coastline, causing damage as far north as Copiapó and as far south as Concepción, which are 1000 km apart. The coastal city of Valparaíso was completely destroyed and Santiago sustained significant damage (Lomnitz, 1970; Udías et al., 2012). Damage from a high (>10 m) tsunami accompanying the AD 1730 earthquake stretched from La Serena in the north to

Concepción in the south. The AD 1730 earthquake was at least as large as the 2010

Maule (Mw 8.8) earthquake but probably did not exceed the size of the 1960 Valdivia (Mw

9.5) earthquake, the two largest historical events in south central Chile (Udías et al.,

2012). The sparse coastal population at the time of the 1730 earthquake resulted in probable undocumented widespread coastal deformation. However, considering that the last three plate-boundary earthquakes in the Valparaíso region produced coastal uplift

(Lomnitz, 1970; Barrientos, 2007), we would expect the AD 1730 earthquake produced similar deformation.

The AD 1822, 1906, and 1985 earthquakes produced significant but more localized damage along the coast. The tsunamis accompanying the AD 1822, 1906, and 1985 events were lower (<4 m) than expected for the magnitude of the earthquakes (Moreno et al., 2007). The three earthquakes produced 0.4-1.2 m of coseismic uplift in and around Quintero (Comte et al., 1986). Thus, along the central Chilean subduction zone,

59 CHAPTER 3 historical earthquakes suggest that the earthquake deformation cycle is characterized by repeated instances of variable coseismic uplift interspersed with gradual interseismic subsidence, and deformation patterns for prehistoric earthquakes were probably at least as variable (Barrientos, 1996, 1997).

3.2.2 Quintero lowland

The Quintero lowland is located about 35 km north of Valparaíso and 70 km north of San

Antonio, two of the largest ports in Chile (Fig. 3.1b, 3.3). The lowland (32.5ºS) consists of a 1.2 km2 coastal plain between 2 and 4 m above modern mean tide level (MTL). The lowland has no fluvial input and is isolated from the sea. Poor drainage and a high water table have allowed freshwater marsh vegetation to grow in a small area (0.3 km2) in the southern lowland otherwise dominated by woodland, shrubland, and scrub vegetation

(Fig. 3.3c; Villa-Martínez and Villagrán, 1997; Maldonado and Villagrán et al., 2006).

Marsh plants Scirpus americanus, S. cernuus, and S. inundatus dominate the lower areas of the lowland, while Sarcocornia fruticulosa, Atriplex prostrata, and

Chenopodium frigidum are found on higher ground (Villa-Martinez and Villagran,

1997). Pleistocene dunes with elevations up to 80 m above MTL border the marsh.

Pleistocene dunes are composed of mica-poor, poorly sorted fine and medium sand. A prograding late Holocene dune sequence (~4 m MTL) extends ~1 km inland, separating the marsh from the current coastline (Caviedes, 1972; Fig. 3.3c). In contrast to the

Pleistocene dunes, the late Holocene dunes and beach sediment consist of mica-rich, well-sorted fine quartz sand with little variation in particle size between summer and winter (Martínez et al., 2011).

60 CHAPTER 3 The modern climate in the Quintero coastal region is semiarid with dry summers and wet winters (280 mm of rainfall per year), although winter droughts are not uncommon

(Villa-Martínez and Villagrán, 1997). The modern Quintero coast is wave dominated and microtidal (Great Diurnal tidal range of ~1.6 m between MLLW-MHHW) as measured by the Chilean Hydrographic and Oceanographic service (SHOA) at Quintero. Two kilometers south of Quintero, the greatest incidence of wave direction is generally coincident with the predominant wind direction (from the W, SW, and NW) with maximum wave heights of 1-3 m (Martínez et al., 2011). No historical tropical cyclones have affected the Chilean coast because atmospheric circulation patterns preclude the formation of such low-pressure storm systems in the eastern South Pacific Ocean

(Fedorov et al., 2010).

Palynology and historical records show the Quintero lowland has undergone significant environmental changes during the Holocene. Changes in pollen assemblages suggest that during the mid Holocene, the landward fringes of the Quintero lowland were part of an estuary and tidal marsh (Fig. 3.3c; e.g., Villa-Martínez and Villagrán, 1997; Maldonado and Villagrán et al., 2006). Assemblages within a sandy silt overlying basal Pleistocene sand indicate that sea level was higher prior to 5000 cal yr BP and that the lowland consisted of tidal flats and beach berms with patchy dune (grasses and small flowering plants) and sparse aquatic vegetation within interdunal depressions. At 4200 cal yr BP, a sharp increase in aquatic vegetation and brackish mollusks in fine silty sediment are consistent with tidal marsh and brackish lagoon environments. After 4000 cal yr BP, pollen assemblages show a decrease in aquatic vegetation and an increase in freshwater marsh vegetation suggesting decreased tidal inundation of the lowland during late-

Holocene sea-level fall (Villa-Martínez and Villagrán, 1997). Since 2000 cal yr BP the

61 CHAPTER 3 vegetation and climate at Quintero were similar to the present (Maldonado and Villagrán et al., 2006). Historical maps from the AD 1700s and 1800s show a lowland environment similar to that of today, but include a lagoon covering parts of the lowland (Fig. 3.3b).

The lagoon has since drained, probably due to human modifications. In general, the upper meter of sediment has undergone significant anthropogenic modification during

19th and 20th century construction projects aimed at preparing the lowland for residential development (never completed), and later a Chilean Air Force airstrip.

3.3 APPROACH AND METHODS

To decipher a history of coseismic land-level change during large earthquakes and tsunami deposition for the Quintero lowland, we described 57 cores, hand-dug pits, and exposures to a depth of 2-3 m (Fig. 3.3c). The Chilean Air Force, which currently occupies the lowland, dug 10- to 50-m-long, ~2-m deep backhoe trenches for us during the construction of a new runway (Fig. 3.4). For coring, we used a 1-m-long, 25-mm- diameter, half-cylinder gouge corer. All sections were photographed and their lithology, color (Munsell Soil Color Chart, 2009), and stratigraphy described in the field. We used a differential GPS system (Leica 1200) to survey the lowland (Fig. 3.5a). All elevations were tied into a benchmark and tidal datum maintained by the SHOA at Quintero, and reported as elevation (in meters) relative to MTL. Multiple transects perpendicular and parallel to the coast were completed and outcrop, core and pit locations were projected onto these transects with less than 15 cm vertical errors.

The most complete stratigraphic sequences were found along the slightly higher fringes of the coastal lowland (Fig. 3.5). We selected two pits from this area (Fig. 3.6; 43 and 18) for detailed analysis; monoliths were collected in the field and subsampled in the

62 CHAPTER 3 laboratory for grain-size distribution, organic content (loss on ignition), diatom analyses and radiocarbon dating.

3.3.1 Grain-size and loss on ignition analyses

For comparison of grain-size distributions among fossil and modern sediment, we collected modern surface (0-1 cm depth) samples from the beach bordering the lowland and the Pleistocene dunes (Fig. 3.7b). We sampled beach sediments in both summer and winter to account for seasonal variations in grain size. In fossil sections, subsampling for grain size and loss on ignition (LOI) was conducted at 1 cm intervals near lithologic contacts and at 5 cm intervals in homogenous sediments.

Modern and fossil grain-size samples were treated with hydrogen peroxide (30%) to remove organic material and analyzed using a Beckman Coulter LS230 laser diffraction grain-size analyzer. Grain-size interpretations are based on the Wentworth Phi Scale and reported as differential volume (the percentage of total volume that each size class occupies). The grain-size distributions were interpolated and gridded using a triangular irregular network (TIN) algorithm (Sambridge et al., 1995) and plotted as color surface plots using Geosoft Oasis TM software (Donato et al., 2009). The mean grain size and

D10 value (diameter at which 10% of a sample's volume comprises smaller particles) of each sample were used to characterize lithologic units.

Fossil samples for LOI (a proxy for the percentage of organic matter) were dried at 105C for 12 hours, weighed, baked at 550C for 6 hours, and weighed again to estimate the weight lost (LOI; Ball, 1964).

63 CHAPTER 3 3.3.2 Diatom analysis

We subsampled the lower stratigraphy of pit 43 (including sand beds F-D) and upper stratigraphy of pit 18 (including sand beds C-A) for diatom analysis (Fig. 3.8, 3.9).

Subsampling intervals ranged from 2 cm near lithologic contacts to 15-cm intervals in homogenous sediment. To remove organic material, we oxidized ~1 g of sediment with nitric acid in a microwave digestion system (Charles et al., 2002). A known volume of digested sample (between 25 and 100 l depending on the diatom concentration) was then pipetted and distributed evenly on a cover slip. The cover slip was dried overnight and then inverted and mounted on a glass slide using Naphrax (Charles et al., 2002). A total of 49 diatom slides were prepared from pits 43 and 18. Diatoms were identified to species level using a light microscope under oil immersion at 1000x magnification with reference to Krammer and Lange-Bertalot (1986, 1988, 1991a,b) and Lange-Bertalot

(2000). When possible, 300 diatoms were identified and counted in slides with each species expressed as a percentage of total diatom valves counted. Fragments containing more than half a valve were included in the count. Only species that exceeded 5% of total valves counted were used for paleoecological interpretation (Horton et al., 2007). We identified 49 species in 19 genera from 49 samples in pits 43 and 18. Diatoms were classified by salinity (marine, brackish, and freshwater) and life-form (planktonic, epipelic, epiphytic, aerophilic) based on the ecological preferences outlined in Chilean

(Rivera, 1983; Rebolledo et al., 2005, 2011) and global (Krammer and Lange-Bertalot,

1986, 1988, 1991a,b; Vos and de Wolf, 1988, 1993; Denys, 1991; Hartely et al., 1996) catalogs.

The depositional environment (high-marsh/upland, brackish lagoon, low-marsh, or tidal flat) represented by each stratigraphic unit was determined based on a qualitative 64 CHAPTER 3 interpretation of the relative abundances of diatoms within salinity and life-form groups.

The marine diatom group includes marine and marine-brackish species that thrive in salinities exceeding 30 practical salinity units (psu). The brackish diatom group includes brackish-marine and brackish species that tolerate salt concentrations between 0.2 and

30 psu. The freshwater group includes fresh-brackish and fresh diatoms that generally occur in salt concentrations less than 0.2 psu. Diatom taxa that live attached to plants are defined as epiphytic forms; taxa that live on wet sediments are defined as epipelic forms; taxa that live on wet sediments but are able to survive temporarily dry conditions are defined as aerophilic forms. Tychoplanktonic diatoms include an array of species that live in the benthos, but are commonly found in the plankton. Diatoms that float in the water column and do not live attached to any substrate are defined as planktonic forms

(Vos and de Wolf, 1988, 1993).

Samples were also scanned for the abundance of phytoliths and chrysophyte cysts to help distinguish freshwater from tidal environments. Phytoliths typical of brackish and fresh coastal grasses were identified to the subfamily level (Pooideae and Bambusoideae subfamilies) with reference to the classifications of Lu and Liu (2003).

To examine the pattern of diatom assemblage variation within and between samples, we used detrended correspondence analysis (DCA; Fig. 3.10). Samples with similar species compositions are grouped together in the DCA bi-plot and samples with statistically different species compositions plot apart (Horton et al., 2007). The percent abundance of grass phytoliths and chrysophyte cysts were included in the DCA analysis.

65 CHAPTER 3 3.3.3 Radiocarbon dating

Where possible, we collected plant macrofossils for radiocarbon dating from organic sediments that gave either limiting maximum ages (usually below each contact) or limiting minimum age (in growth position above each contact) to constrain the age of the sand beds in Pits 43 and 18 (Table 3.1). Following Kemp et al. (2013), we collected samples that were in growth position (rhizomes) or so delicate (e.g., seeds, leaf parts) that they would probably have decayed or been broken by transport soon after death, thus reducing the likelihood of dating material much older than contact burial.

Radiocarbon ages were calibrated using OxCal radiocarbon calibration software (Bronk

Ramsey, 2009) with the SHCal13 data set of Hogg et al. (2013). Calibrated age ranges are shown at two standard deviations where years ‘before present’ (BP) is years before AD

1950. We use maximum and minimum limiting radiocarbon ages to calculate age probability distributions for inferred tsunami beds from pits 43 and 18 with the V- sequence feature of OxCal (Fig. 3.11; Bronk Ramsey 2008; e.g., DuRoss et al., 2011;

Berryman et al., 2012). To estimate inter-event times and recurrence intervals of paleoearthquakes and tsunamis we used the “difference” function in OxCal.

3.4 RESULTS

We observed five main lithostratigraphic units (units 1, 2, 3, 4, and 5) and six distinct sand beds (beds A, B, C, D, E, and F) that occur within units 2, 3, and 4 (Fig. 3.5b). A compact grey (5YR 7/1) basal sand containing very little organic material uniformly underlies the lowland between 1 and 2.5 m MTL (unit 1). Landward of 1650 m from the present coastline, a dark brown (10YR 2/1) organic sand (unit 2) overlies the basal sand

(cf. pit 43). The organic sand is ~1.5 m thick and contains visible (5-10 mm long) herbaceous plant fragments. In landward locations between 1100 m and 1650 m, a light 66 CHAPTER 3 brown (10YR 5/3) organic silt (unit 3) overlies the basal sand. The unit is ~0.5 m thick and contains humified organic matter, but few identifiable herbaceous plant fragments.

A dark brown (10YR 2/1) organic silt (unit 4) that is ~1 m thick overlies unit 3. It is distinguished from unit 3 by its darker color and the presence of visible (5-10 mm long) herbaceous plant fragments. Above the organic sands and silts is ~1 m of brown (10YR

4/3) reworked dune sand containing little organic material (unit 5). In landward locations the source of the sand is probably the adjacent Pleistocene dunes, whereas in seaward locations up to 1100 m from the present coastline, the source of the sand is likely the modern <1000-yr-old coastal dunes.

Six ~4-cm-thick sand beds are interbedded within the organic silt and sand (units 2, 3, and 4) and extend up to 200 m across the lowland. We labeled the sand beds from oldest

(F) to youngest (A). All six sand beds can be correlated throughout unit 2. Units 3 and 4 only contain the three youngest beds (beds C, B, and A). Three thinner, discontinuous sand beds were found in more seaward sections (Fig. 3.5b). However, because we were not able to correlate these beds between sections and we are unsure of their origin, they will not be discussed further. Beds F-A were distinct from their host sediments due to their sharp (<1-3 mm) upper and lower contacts, lack of organic matter, uniform thickness, lateral extent, and planktonic marine fossils. Erosive contacts and rip-up clasts were observed in beds F, E, and D.

3.4.1 Basal sand

The basal sediments of pits 43 (2.10 to 2.20 m MTL) and 18 (1.40 to 1.50 m MTL) are composed of grey (5YR 7/1), well sorted, mica-rich, very fine quartz sand (unit 1; mean =

2.45 Φ; D10 = 4.38 Φ; Fig. 3.6a) and contain little organic material (LOI < 10%). Marine

67 CHAPTER 3 and brackish epiphytes (e.g. Achnanthes brevipes and Tabularia fasciculata) dominate basal sand diatom assemblages in both pits (Fig. 3.8, 3.9). Pit 18 contains additional marine and brackish species including the marine epiphyte Cocconeis scutellum, the marine planktonic Thalassiosira lacustris, the brackish epipelon Amphora coffeaeformis, and the brackish tychoplanktonic Cyclotella Meneghiniana (Table 3.2).

In Pit 18, a fragile twig with leaf scars collected from the upper contact of the basal sand yields a maximum limiting age of 5655-5896 cal yr BP for the onset of Holocene sediment deposition in this portion of the lowland (Fig. 3.6b).

3.4.2 Sand bed F and under-and overlying units (Pit 43)

Between 2.20 and 2.28 m MTL, a dark brown, poorly sorted organic sand (unit 2; mean

= 2.36 Φ; D10 = 4.96 Φ) overlies the basal sand in Pit 43 (Fig. 3.6a, 3.7a). The unit contains more organic material (LOI 25%) than the underlying basal sand, but diatom assemblages are similar, with an assemblage dominated by the marine diatoms A. brevipes and T. fasciculata (Fig. 3.8).

At 2.28 m MTL, an erosive contact separates unit 2 from bed F, a 4-cm-thick, tan, well sorted, mica-rich, fine quartz sand (mean = 2.26 Φ; D10 = 3.22 Φ) containing little organic material (LOI < 10%). The sand bed contains rip-up clasts from the underlying unit. Although bed F contains a marine and brackish diatom assemblage similar to that of the underlying sediment, there is an increase in the marine planktonic T. lacustris. A decreased fine fraction—shown by a decrease in D10 value—distinguishes the grain size of bed F from under- and overlying sandy units. Four grain-size samples taken at 1 cm vertical intervals show subtle upward fining and an increased fine fraction towards the top of the bed (Fig. 3.7a). 68 CHAPTER 3 A dark brown, poorly sorted organic (LOI 28%) sand (unit 2; mean = 2.54 Φ; D10 = 7.23

Φ) overlies bed F at 2.33 m MTL. Diatom assemblages within this organic sand are similar to assemblages underlying bed F. But there is an increase in the freshwater epiphyte Rhopalodia gibberula and grass phytoliths and chrysophyte cysts within the overlying unit (Fig. 3.8). The DCA reflects this increase in freshwater taxa with a subtle change in DCA sample scores between units below and above the bed (Figure 3.10a).

A rhizome collected from the organic sand immediately overlying bed F yields a minimum limiting age for sand deposition of 5931-6182 cal yr BP (Fig. 3.6a).

3.4.3 Sand bed E and under-and overlying units (Pit 43)

At 2.38 m MTL, an erosive contact separates unit 2 from bed E, a 3-cm-thick, tan, well sorted, mica-rich, fine quartz sand (mean = 2.23 Φ; D10 = 3.10 Φ; Fig 3.6a, 3.7a) containing little organic material (LOI 8%). The sand bed contains rip-up clasts from the underlying unit. Diatom assemblages within bed E are dominated by marine and brackish species, including planktonic species (e.g., T. lacustris and Cyclotella meneghiniana) not found in units under- and overlying the bed. Dominant marine and brackish diatoms in the lower 1 cm of bed E include A. brevipes, A. coffeaeformis, and

Cocconeis scutellum. In the upper centimeter of bed E, A. coffeaeformis, T. fasciculata, and the brackish epipelon Gyrosigma nodiferum increase in abundance (Fig. 3.8). A decreased fine fraction distinguishes the grain size of bed E from under- and overlying sandy units. Three grain-size samples taken at 1 cm vertical intervals show subtle upward fining and an increased fine fraction towards the top of the sand bed (Fig. 3.7a).

A dark brown, poorly sorted organic (LOI 32%) sand (unit 2; mean = 2.28 Φ; D10 = 3.46

Φ) overlies bed E at 2.44 m MTL. Within this organic sand, there is a notable increase in 69 CHAPTER 3 freshwater diatoms (e.g., Navicula tripuncta, Nitzschia lanceolata, Nitzschia linearis, R. gibberula and Staurosirella pinnata) compared to the marine dominated diatom assemblage in sediments underlying bed E. The significant increase in freshwater diatom diversity and abundance above the sand bed is reflected by a shift in DCA sample scores from marine to freshwater dominated samples (Figure 3.10a).

A rhizome collected from the organic sand immediately underlying bed E yields a maximum limiting age for sand deposition of 5492-5732 cal yr BP (Fig. 3.6a).

3.4.4 Sand bed D and under-and overlying units (Pit 43)

The diatom assemblage in the dark brown, poorly sorted organic sand (LOI 29%) immediately underlying bed D at 2.4 m MTL is dominated by the brackish diatom A. coffeaeformis (85%). At 2.5 m MTL, an erosive contact separates unit 2 from bed D, a 7- cm-thick, tan, well sorted, mica-rich, fine quartz sand (mean = 2.20 Φ; D10 = 3.15 Φ; Fig.

3.6a, 3.7a) containing little organic material (LOI < 5%). The sand bed contains rip-up clasts from the underlying unit. The diatom assemblage within bed D is dominated by marine and brackish species (e.g., A. brevipes, A. coffeaeformis, T. fasciculata, and

Planothidium delicatulum), including the marine planktonic T. lacustris, not found in the organic units under and overlying the bed (Fig. 3.8). A. brevipes is abundant in the lower 3 cm of bed D and decreases in the top 4 cm of the deposit while A. coffeaeformis increases. A decreased fine fraction distinguishes the grain size of bed D from under- and overlying organic sandy units. Seven grain-size samples taken at 1 cm vertical intervals show subtle upward fining in the bed (Fig. 3.7a).

A dark brown, poorly sorted organic (LOI 35%) sand (unit 2; mean = 2.36 Φ; D10 = 4.85

Φ) overlies bed D at 2.56 m MTL. Within this organic sand, there is an increase in 70 CHAPTER 3 freshwater (e.g., R. gibberula, Navicula rhynchocephala, and Denticula elegans), brackish (G. nodiferum), and marine (e.g. A. brevipes, T. fasciculata) diatoms and a significant decrease in the brackish A. coffeaeformis compared to the brackish dominated assemblage in sediments underlying bed D. DCA sample scores above bed D reflect the significant change in species composition (Figure 3.10a). Similar diatom assemblages are found in the organic unit up to 2.79 m MTL, with small increases in freshwater species observed at 2.7 m MTL.

A rhizome collected from the organic sand above bed D yields a minimum limiting age for sand deposition of 4856-5260 cal yr BP (Fig. 3.6a).

3.4.5 Sand bed C and under-and overlying units (Pit 18)

Between 1.5 and 1.89 m MTL, a light brown, very poorly sorted organic (LOI 37%) silt

(unit 3; mean = 3.41 Φ; D10 = 6.88 Φ; Fig 3.6b, 3.7a) overlies the basal sand in Pit 18.

Like the underlying basal sand, diatom assemblages in the organic silt are dominated by marine (e.g., T. fasciculata, A. brevipes var. intermedia, A. brevipes, and Cocconeis scutellum) and brackish (e.g., G. nodiferum and Surirella striatula) species (Fig. 3.9).

At 1.89 m MTL, a sharp contact separates unit 3 from bed C, a 4-cm-thick, grey, well sorted, mica-rich, fine quartz sand (mean = 2.45 Φ; D10 = 3.77 Φ; Fig. 3.6b, 3.7a) containing little organic material (LOI < 6%). The diatom assemblage within bed C is dominated by marine and brackish diatoms (e.g., T. fasciculata, A. brevipes var. intermedia, A. brevipes, S. striatula and G. nodiferum) including the planktonic Paralia sulcata, which is not found in organic units under- and overlying the sand bed (Fig. 3.9).

An increase in mean grain size and a decreased fine fraction distinguishes bed C from under- and overlying silty sediments by (Fig. 3.7a). 71 CHAPTER 3 A light brown, very poorly sorted organic (LOI 40%) silt (unit 3; mean = 3.20 Φ; D10 =

7.13 Φ) overlies bed C at 1.93 m MTL. Within this organic silt, there is an increase in freshwater and fresh-brackish diatoms (e.g., Nitzschia amphibia, Gomphonema parvulum, and Nitzschia frustulum) and grass phytoliths and chrysophyte cysts compared to the marine dominated assemblage in sediments underlying bed C. The significant increase in freshwater diatom diversity and abundance above the bed is reflected by a shift in DCA sample scores from marine to freshwater dominated samples

(Figure 3.10b).

A rhizome and a piece of charcoal collected from the organic silt immediately below bed

C yield statistically indistinguishable maximum limiting ages of sand deposition (4297-

4529 and 4160-4509 cal yr BP). The pooled mean of the ages is 4291-4517 cal yr BP (Fig.

3.6b).

3.4.6 Sand bed B and under-and overlying units (Pit 18)

The light brown organic silt above bed C is replaced by a darker brown (LOI 50%), very poorly sorted organic silt at 2.08 m MTL (unit 4; mean = 3.60 Φ; D10 = 6.99 Φ; Fig.

3.6b, 3.7a). The diatom assemblage within the dark brown organic silt is dominated by the brackish diatom Diploneis smithii. At 2.24 m MTL, a sharp contact separates unit 4 from bed B, a 3-cm-thick, brown, well sorted, mica-rich, fine quartz sand (mean = 2.65

Φ; D10 = 4.08 Φ; Fig. 3.6b, 3.7a) containing little organic material (LOI < 5%). Bed B contains a mixed diatom assemblage including marine (e.g., T. fasciculata), brackish

(e.g., S. striatula) and freshwater (e.g., D. elegans, R. gibberula, and D. smithii) species

(Fig. 3.9). The marine and brackish planktonics P. sulcata and T. lacustris, not found in under- and overlying organic units, are observed in bed B. An increase in mean grain size

72 CHAPTER 3 and a decreased fine fraction distinguishes bed B from under- and overlying silty sediments (Fig. 3.7a).

A dark brown very poorly sorted, organic (LOI 51%) silt (unit 4; mean = 3.83 Φ; D10 =

7.17 Φ) overlies bed B at 2.27 m MTL. Within this organic silt, there is an increase in freshwater diatoms (e.g., R. gibberula, D. elegans, and Diploneis subovalis) and an increase in grass phytoliths and chrysophyte cysts compared to the brackish dominated assemblage in sediments underlying bed B. The significant increase in freshwater diatom diversity and abundance above the sand bed is reflected by a shift in DCA sample scores from brackish to freshwater dominated samples (Fig. 3.10b).

A rhizome collected from the organic silt immediately underlying bed B yields a maximum limiting age for sand bed deposition of 3711-3974 cal yr BP (Fig. 3.6b).

3.4.7 Sand bed A and under-and overlying units (Pit 18)

At 2.35 m MTL, the brackish diatom D. smithii (Fig. 3.9) increases in abundance and remains the dominant species within the dark brown organic silt until 2.45 m MTL, when the unit is sharply overlain by bed A, a 3-cm-thick, organic rich (LOI 48%), well sorted, mica-rich, fine quartz sand (mean = 2.56 Φ; D10 = 4.44 Φ; Fig. 3.6b, 3.7a).

Diatom assemblages within bed A are dominated by marine species (e.g., T. fasciculata) including the marine planktonic diatoms T. lacustris and Thalassiosira weissflogii, not found in organic units under- and overlying the sand bed (Fig. 3.9). An increase in mean grain size and a decreased fine fraction distinguishes bed A from under- and overlying silty sediments. Bed A shows upward fining and an increased fine fraction towards the top of the bed (Fig. 3.7a).

73 CHAPTER 3 A dark brown, very poorly sorted organic (LOI 45%) silt (unit 4; mean = 4.86 Φ; D10 =

8.31 Φ) overlies bed A at 2.48 m MTL. Within this organic silt, there is an increase in freshwater diatoms (e.g., D. elegans, P. brevicostata, and N. amphibia) and grass phytoliths and chrysophyte cysts compared to the brackish dominated assemblage in sediments underlying bed A. The significant increase in freshwater diatom diversity and abundance above the bed is reflected by a shift in DCA sample scores from brackish to freshwater dominated samples (Fig. 3.10b). Diatom assemblages for the remainder of Pit

18 are dominated by freshwater species.

A rhizome collected from the organic silt immediately underlying bed A yields a maximum limiting age for sand bed deposition of 3567-3821 cal yr BP (Fig. 3.6b).

3.5 DISCUSSION

Based on stratigraphy, lithology (grain-size distribution and LOI), diatom, and radiocarbon analyses, we infer the depositional environment represented by each lithologic unit and the timing of environmental changes. In the ~6000 years represented by the lithofacies in our most complete stratigraphic sections, diatom assemblages indicate that the geomorphology of the Quintero lowland was significantly different than today. Diatoms show the landward fringes (where pits 43 and 18 are located) of the lowland transitioned from a mostly marine and brackish environment in the early

Holocene to a freshwater marsh environment by the mid to late Holocene (Fig. 3.8, 3.9).

Within this long-term freshening trend, diatoms fluctuate back and forth between marine and/or brackish dominated assemblages and brackish and/or freshwater assemblages punctuated by widespread, anomalous sand beds, which we interpret below to have been deposited by tsunamis. We infer that the unusually sudden and long lasting

74 CHAPTER 3 changes in diatom assemblages associated with five of the six sand beds signal sudden

RSL falls during times of decimeter-scale coseismic uplift.

3.5.1 Evidence for tsunami deposition

We infer a tsunami origin for the six sand beds within the Quintero lowland sequence.

We base our interpretation on characteristics shared by all six beds including: significant lateral extent, uniform thickness, sharp (1-3 mm) lower and upper contacts, mixed diatom assemblages that include anomalous marine planktonic species, mica content, and upward fining sequences. Such characteristics have been found in modern tsunami deposits from Papua New Guinea (e.g., Gelfenbaum and Jaffe, 2003; Dawson, 2007),

Thailand (e.g., Jankaew et al., 2008; Sawai et al., 2008), Sumatra (e.g., Monecke et al.,

2008), Chile (e.g., Cisternas et al., 2005; Horton et al., 2011; Garrett et al., 2013; Ely et al., 2014), and Japan (e.g., Goto et al., 2011; Szczuciński et al., 2012). Beds F-A can be correlated over distances of 100 to 200 m. Radiocarbon ages from plant macrofossils underlying bed C in pits 18 and 43 are statistically indistinguishable, consistent with lateral continuity of this bed. Beds F, E, and D have erosive lower contacts and sharp upper contacts and contain rip-up clasts of material reworked during inferred high- energy, turbulent flows (e.g., Goff et al., 2012). Beds C, B, and A also display sharp upper and lower contacts but lack rip-up clasts. The lateral continuity and tabular geometry of the sand beds rule out localized processes, such as tidal channel migration, liquefaction, or sand venting during earthquakes as the process responsible for sand bed deposition

(e.g., Atwater, 1992; Allen, 2000; Williams et al., 2005; Arcos et al., 2012).

We infer that the marine, brackish, and freshwater diatoms in all six sand beds reflect tsunami surges and subsequent backflows across subtidal, intertidal and supratidal

75 CHAPTER 3 environments of the lowland (e.g., Hawkes et al., 2007; Sawai et al., 2009; Horton et al.,

2011; Szczuciński et al., 2012). The presence of anomalous marine and brackish planktonic species (e.g., T. weissflogii, T. lacustris, C. meneghiniana, and P. sulcata) in the sand beds supports a tidal-flat or subtidal source for the sediment and eliminates a fluvial source (Fig. 3.12b, 3.13b; e.g., Hemphill-Haley, 1996; Sawai et al., 2008). Our analysis of likely paleo-sediment sources shows that the sand in the inferred tsunami beds was mostly derived from beach and dune erosion. Possible sediment sources for the inferred tsunami beds include the Pleistocene dunes bordering the lowland, the basal tidal flat sand underlying the lowland, and the late Holocene dune, beach, and littoral sand. The Pleistocene dune sand is composed of brown, poorly sorted, fine and medium sand with no mica (mean = 1.87 Φ; D10 = 2.56 Φ; Fig. 3.7b), whereas the basal tidal flat sand and late Holocene dune, beach, and littoral sands are composed of tan or grey, well sorted, mica-rich, very fine and fine quartz sand (mean = 2.45 Φ; D10 = 4.38 Φ). All inferred tsunami beds display similar characteristics to the basal tidal flat sand and late

Holocene coastal sediment; beds are composed of well sorted, mica-rich, fine quartz sand (Fig. 3.7b). The physical characteristics of the sand beds and the presence of mica, which is not found in the Pleistocene dunes, make aeolian deposition unlikely (e.g.,

Sawai et al., 2012).

All but one sand bed (bed B is the exception) drapes preexisting paleo-topography and displays upward fining and an increased fine fraction towards the top of the bed. These features are indicative of sediment settling out of suspension as a tsunami flow decelerates after reaching its inland extent (Fig. 3.7a; Jaffe and Gelfenbaum et al., 2007;

Morton et al., 2007; Jaffe et al., 2011). Diatom analyses also reveal grading of diatom valves, a feature interpreted to reflect variable flow speed in the 2004 Indian Ocean

76 CHAPTER 3 tsunami and 2010 south-central Chile tsunami (Sawai et al., 2009; Horton et al., 2011).

In sands F, E, and D, larger diatoms such as A. brevipes (>50 m; thick, oval shaped valve) were abundant at the bottom of the sand beds while silt-sized diatoms such as A. coffeaeformis (<30 m; narrow valve) increased in abundance towards the top of the beds.

3.5.2 Evidence for sudden, widespread, and lasting uplift

Our diatom results show sudden, widespread, and lasting RSL fall, which we infer is caused by coseismic uplift, associated with deposition of five of the six tsunami sands.

These inferences are consistent with previous paleoseismic studies that used diatoms to support lithologic inferences of coseismic uplift (e.g., Sawai, 2001; Shennan et al., 2009;

Briggs et al., 2014). However, unlike most previous studies, we observed no major change in lithology coincident with diatom evidence for uplift. We found minimal variation in the grain size and organic content (<20% variation) of units under- and overlying the tsunami-deposited sand beds (Fig. 3.6, 3.7). This result is consistent with studies of modern clastic-dominated marshes that display less than a 25% difference in

LOI and little change in lithology between the low and high marsh (Horton, 1999;

Hawkes et al., 2010). Therefore, sudden uplift causing a shift from a low-marsh to a high-marsh environment would not necessarily result in a noticeable change in grain size or organic content, but might be evident in diatom assemblages due to reduced tidal inundation.

We observed an increase in siliceous freshwater microfossils in units overlying beds F, E,

C, B, and A (Fig. 3.12a, 3.13a), suggesting a fall in RSL consistent with coseismic uplift.

DCA results support our qualitative interpretations of sudden changes in diatom species

77 CHAPTER 3 composition (Fig. 3.10a, 3.10b). Units underlying the sand beds generally contain diverse marine and brackish diatom assemblages dominated by tidal flat epiphytes (A. brevipes,

T. fasciculata, C. scutellum) and/or brackish species such as the epipelic A. coffeaeformis and D. smithii, common in shallow estuarine conditions (Lepland et al.,

1995; Chagué-Goff et al., 2002). Units overlying sand beds display an increase in aerophilic freshwater epiphytic diatoms (e.g., R. gibberula, D. elegans, and G. parvulum) and grass phytoliths and chrysophyte cysts indicating fresher, drier conditions with increased density of vegetation (e.g., Rühland et al., 2000; Fig. 3.10a,

3.10b). Above bed D, there is an increase in both freshwater and marine species and a slight decrease in grass phytoliths and chrysophyte cysts. Thus the evidence for uplift across this contact is equivocal.

The diatom assemblages in organic units overlying sand beds display a lasting shift in paleoecology, rather than a temporary response to short-term sea-level fluctuations (e.g.,

El Niño events; Witter et al., 2001) or freshwater inundation of the lowland (Hemphill-

Haley, 1995a). The influx of freshwater species in the units overlying sand beds is followed by a gradual return to brackish conditions (e.g., increases in A. coffeaeformis below bed D and increases in D. smithii below beds B and A) and a decrease in freshwater species, suggesting an RSL rise, and hence interseismic land-level fall (Fig.

3.8, 3.9).

On the basis of shifts in diatom assemblages in lowland units under- and overlying sand beds, we estimate the amount of coseismic uplift across each contact. Our estimates are only semi-quantitative because of the lack of suitable modern analog diatom data from tidal environments in central Chile. We use the modern tidal range at Quintero and the

78 CHAPTER 3 typical position within the tidal frame in which our diatom species are found to make the estimates (e.g., Nelson and Kashima, 1993; Hemphill-Haley, 1995a; Witter et al., 2003).

For example, coincident with the deposition of beds F and C, diatoms indicate a shift from a tidal flat positioned below MTL to a low marsh positioned between mean lower high water (MLHW) and mean higher high water (MHHW). At Quintero, MLHW is 0.3 m MTL and MHHW is 0.6 m MTL, indicating a fall in RSL of at least 0.3 m. Coincident with the deposition of bed E, diatoms reflect a shift from a low marsh to a high marsh positioned between MHHW and highest astronomical tide (HAT, 1.1 m MTL), indicating a fall in RSL ranging from a minimum of 0.1 m to a maximum of 0.8 m. Coincident with the deposition of beds B and A, diatoms indicate a shift from a brackish lagoon positioned between MLHW and MHHW and a freshwater marsh positioned at or near

HAT, indicating a fall in RSL of at least 0.5 m.

Our identification of multiple RSL falls, which we infer were caused by coseismic uplift, coincident with deposition of sand beds strengthens our argument for their deposition by high tsunamis accompanying large, subduction-zone earthquakes. Although high storm surges with barrier breaches can produce widespread sand beds with some characteristics similar to tsunami deposits, breach deposits are not coincident with coseismic uplift (Witter et al., 2001; Morton et al., 2007; Switzer et al., 2008; Williams,

2009). Any reconfiguration of lowland geomorphology during storm wave erosion, such as widening of an outlet channel or migration of a spit, would likely increase the marine influence in units above sand beds rather than freshen them (e.g., Sawai et al., 2002).

Tropical cyclone storm surges capable of transporting large amounts of sediment during a single storm are unprecedented in Chile (Fedorov et al., 2010).

79 CHAPTER 3 3.5.3 Relative sea level and the limits to preservation of earthquake and tsunami evidence

The ~6200-3600 cal yr BP record of earthquakes and tsunamis at Quintero coincides with a period of gradual RSL rise shown in ICE-5G VM5a model predictions between

~8000 and ~4000 yrs BP (e.g., Mitrovia and Milne, 2002; Peltier, 2004; Milne et al.,

2005). During this period, the eustatic contribution to sea level was dominant and net sea-level rise created a tidal flat and tidal marsh environment in the Quintero lowland that provided the accommodation space necessary to preserve earthquake and tsunami evidence (e.g., Dura et al., 2011; Brill et al., 2013). The ICE-5G VM5a model suggests this

RSL rise culminated in a highstand of ~1.4 m at ~4200 cal yr BP (Peltier and

Drummond, 2008). Such modeling is consistent with coastal records from central Chile that suggest a RSL highstand of at least 2 m between 6500 and 4000 cal yr BP (Ota and

Paskoff, 1995; Villa-Martínez and Villagrán, 1997; Encinas et al., 2006; Saillard et al.,

2009).

We found no widespread stratigraphic evidence of historical earthquakes or tsunamis in the younger, sandier, seaward sections the Quintero lowland (Fig. 3.5b). Even the AD

1730 tsunami, which recent models suggest exceeded 6 m in areas within 500 m of the

Quintero coast (SHOA model, 2012), left no compelling sedimentary evidence (Lomnitz,

2004). We attribute the scarce evidence elsewhere along the coast to the net emergence of the central Chile coast during the late Holocene. Net Holocene emergence limits the accommodation space necessary for the preservation of earthquake and tsunami evidence resulting in brief or fragmentary paleoseismic records in Chile (e.g., Atwater et al., 1992; Nelson et al., 2009) and Sumatra (e.g., Dura et al., 2011; Grand Pre et al.,

2012). For example, Nelson et al. (2009) documented stratigraphic evidence of 80 CHAPTER 3 subsidence in the Valdivia estuary (39.5S) from the AD 1960 earthquake, but found that a distinct unconformity separated evidence of this event from two previous subsidence events dated to ~1500 and 2300 cal yr BP. Nelson et al. (2009) attributed the fragmentary record of earthquake evidence at Valdivia to erosion and mixing of late

Holocene sediments during net Holocene emergence. In contrast, paleoseismic studies conducted at Cascadia and Alaska, where a combination of tectonic, isostatic, and eustatic sea-level changes has resulted in net Holocene RSL rise with ample accommodation space, document a much more continuous stratigraphic record of coseismic subsidence over as much as 6000 years (Nelson, 2012, figure 16; e.g., Kelsey et al., 2002; Witter et al., 2003; Shennan et al., 2009).

3.5.4 A 500-year-interval between great earthquakes provides context for the earthquake record

The recurrence interval of prehistoric earthquakes and tsunamis at Quintero is comparable to recurrence intervals at other subduction zones thought to preserve evidence of only unusually large (> Mw 8.5) earthquakes and tsunamis. Between 6200 and 3600 yr BP, the interval between great subduction-zone earthquakes at Quintero ranged from ~200 to ~650 years with an average recurrence interval of ~500 years (Fig.

3.11). Similar recurrence intervals have been found for great earthquakes at Cascadia

(e.g., Witter et al., 2003, Nelson et al., 2006), northern Japan (e.g., Sawai et al., 2009), and south central Chile (e.g., Cisternas et al., 2005; Moernaut et al., 2014). The recurrence interval for the three most recent historical earthquakes (AD 1822, 1906, and

1985) in central Chile is much shorter (~80 years) suggesting that the events recorded in the Quintero lowland were unlike those of the historical record.

81 CHAPTER 3 In the mid-Holocene, when GIA models indicate sea level at Quintero was at least 1.4 m higher, it is possible that the lowland was in a position to record uplift from earthquakes comparable to historical events. The <1 m of uplift that characterizes the most recent historical ruptures is consistent with the range of uplift preserved in the Quintero sequence. But the low tsunamis (<4 m) produced by the AD 1822, 1906, and 1985 earthquakes are inconsistent with the distinctive sand beds we discovered in the

Quintero lowland. A better historical analog for our prehistoric earthquakes and tsunamis may be the unusually large AD 1730 earthquake and high tsunami. Uplift estimates for the 1730 earthquake are unknown, but based on the comparable 1960 and

2010 earthquakes in south-central Chile, uplift associated with unusually large earthquakes is variable (0-2 m; Plafker and Savage, 1970). Thus, we interpret the record of coastal uplift and distinctive, widespread sand beds at Quintero as an archive of AD

1730-style earthquakes and tsunamis.

3.6 CONCLUSIONS

We used stratigraphy, lithology, diatom micropaleontology and radiocarbon dating to identify six instances of tsunami inundation at Quintero between 6200 and 3600 cal yrs

BP, five of which were coincident with coastal uplift. Widespread anomalous sand beds display physical characteristics and diatom assemblages indicative of an oceanward source and high-energy deposition. The coincidence of uplift with sand deposition for five of the six sand beds is consistent with a tsunamigenic earthquake as the source for the sand. In the absence of lithologic changes across units above and below the sand beds, an abrupt increase in siliceous freshwater microfossils in units immediately overlying sand beds reveals a sudden fall in RSL coincident with tsunami deposition. The coincidence of uplift with sand deposition helps rule out storm surges and localized

82 CHAPTER 3 coastal processes as mechanisms for sand bed deposition. Paleoecological changes spanning sand beds are laterally continuous over hundreds of meters, indicating large portions of the lowland were affected, ruling out gradual RSL change and estuarine reconfiguration as mechanisms for paleoecological changes coincident with sand deposition. A significant and sustained increase in freshwater influence was observed in units overlying sand beds, followed by a gradual return to brackish and marine conditions, consistent with coseismic uplift and subsequent interseismic subsidence.

Evidence of interseismic subsidence helps rule out gradual emergence of the lowland as a mechanism for paleoecological changes coincident with sand deposition. We infer <1 m of uplift during five of the six earthquakes based on present-day tidal datums and changes in diatom assemblages in units above and below sand beds.

Our interpretations have important implications for evaluating earthquake hazards in central Chile. Basing hazard assessments on only the most recent, well-documented, but smaller earthquakes and tsunamis in 1822, 1906, and 1985, overlooks the higher hazard posed by larger earthquakes and higher tsunamis, such as those which occurred in 1730 and hundreds of years apart between 6200 and 3600 cal yr BP. The recent experience of underestimating the potential of the Sumatra and Japan megathrusts to produce unusually large earthquakes (Jankaew et al., 2008; Sawai et al., 2012)—based on only a limited historical record—emphasizes the importance of using the prehistoric record in estimating the hazard to which the current coastal population is exposed.

Comprehensive earthquake hazard assessments in central Chile should consider the effects of the largest earthquakes and tsunamis that have repeatedly struck this coast.

83 CHAPTER 3 3.7 ACKNOWLEDGMENTS

This work was supported by Chile’s Fondo Nacional de Desarrollo Científico y

Tecnológico (FONDECYT N° 1110848 to MC) and the National Science Foundation

(grant EAR-1144537). Nelson is supported by the Earthquake Hazards Program of the

U.S. Geological Survey. We are grateful to Brian Atwater, Diego Muñoz, Marcelo Lagos, and Yuki Sawai for support during fieldwork, Daniel Melnick for providing elevation data, Rich Briggs for providing a DEM of Chile, Lorena Rebolledo for her preliminary diatom analyses, and Sonja Hausmann and Don Charles at the Academy of Natural

Sciences in Philadelphia, USA, for their assistance with diatom analyses. We thank

Harvey Kelsey, Brian Sherrod, and Simon Engelhart for constructive reviews that substantially improved the manuscript. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government. This paper is a contribution to the IGCP project 588 ‘Preparing for coastal change’.

84 CHAPTER 3

Figure 3.1: Index maps. a) Plate-tectonic setting of Chile in western South America. b) Location of the study area in central Chile, seismic segments, main tectonic features, and estimated rupture lengths of the largest historical earthquakes in central Chile since 1730 compiled from Lomnitz (1970), Kelleher (1972), Comte et al. (1986), Beck et al. (1998), and Melnick et al. (2009). The 1960 and 2010 ruptures are shown for comparison.

85 CHAPTER 3

Figure 3.2: Schematic drawing of coseismic uplift and tsunami inundation in central Chile (modified from Atwater and Hemphill‐ Haley, 1997).

86 CHAPTER 3

Figure 3.3: a) Quintero Bay is located about 35 km north of the port city of Valparaíso. The city of Quintero and the Quintero lowland are located along the southern margin of the bay. b) Map of the Quintero lowland from AD 1876 showing a shallow lagoon, marsh, and forested areas covering the Quintero lowland. The boundary of the lowland (as shown in c) is marked by a black dotted line. c) Satellite Google Earth Pro, 2014 DigitalGlobe imagery of the Quintero lowland with a 10-m-contour topographic overlay. Described cores were projected onto three elevation transects, shown by blue dashed lines. The lowland is bordered by Pleistocene dunes to the south and a prograding dune sequence separates the marsh from the coast. The boundaries of the lowland are shown by a black dotted line. The location of the pollen core is approximate.

87 CHAPTER 3

Figure 3.4: Photographs and map showing an outcrop along the Z-Z′ elevation transect. Sand beds F-A are labeled.

88 CHAPTER 3

Figure 3.5: a) Coast perpendicular elevation profile along transect Y‐ Y′ (see Figure 3.3c for profile location). Elevations are relative to mean tidal level (MTL). Modern tidal datums mean lower high water (MLHW; 0.3 m), mean higher high water (MHHW; 0.6 m), and highest astronomical tide (HAT; 1.1 m) are shown on right. b) Detail of the Quintero lowland stratigraphy showing pit locations and radiocarbon sampling locations. Sands F-A are labeled. The extents of sand beds are plotted above the elevation profile. Pits chosen for detailed analysis (43 and 18) are highlighted in red.

89 CHAPTER 3

Figure 3.6: a, b) High resolution grain-size distribution color surface plots (Φ) showing the differential volume of sediments (the percentage of total volume that each size class occupies), detailed lithology, D10 values (diameter at which 10% of a sample's mass comprises smaller grains), and LOI (weight %) for stratigraphic columns of pits 43 and 18. Elevations are relative to mean tide level (MTL). Radiocarbon sampling locations and calibrated radiocarbon ages are plotted on columns. Radiocarbon sample numbers correspond to sample numbers in Table 3.1.

90 CHAPTER 3

91 CHAPTER 3 Figure 3.7 (previous page): a) Grain-size differential volume curves for stratigraphic units and sand beds F- A. Lines connecting differential volume curves show normal grading in sand beds. b) Grain size differential volume curves for modern surface sediments of the Quintero lowland.

Figure 3.8: Summary of microfossil analyses for pit 43. Triangle symbols on stratigraphic column show locations of numbered samples. Relative abundance of diatoms (showing only those species that are ecologically diagnostic) is expressed as a percent of the total count. Abundance of grass phytoliths and chrysophyte cysts expressed as a percent of total siliceous microfossils.

92 CHAPTER 3

Figure 3.9: Summary of microfossil analyses for pit 18. Triangle symbols on stratigraphic column show locations of numbered samples. Relative abundance of diatoms (showing only those species that are ecologically diagnostic) is expressed as a percent of the total count. Abundance of grass phytoliths and chrysophyte cysts expressed as a percent of total siliceous microfossils.

93 CHAPTER 3

Figure 3.10: Results of detrended correspondence analysis (DCA) for diatom samples from Pit 43 (a) and Pit 18 (b). Samples with similar species compositions are grouped together in the DCA bi-plot and samples with statistically different species compositions plot apart (Horton et al., 2007). The percent abundance of grass phytoliths and chrysophyte cysts are included in the DCA analysis. Arrows show how assemblage bi- plot scores change from below to above labeled bed contacts.

94 CHAPTER 3

Figure 3.11: Age probability distributions (brackets show 2 interval) for inferred tsunami-deposited beds from pits 43 and 18 calculated using the V-sequence feature of OxCal. Distributions limited by a maximum age are purple; those limited by a minimum age are blue (Table 3.1). The average recurrence interval of inferred tsunamis is ~500 years.

95 CHAPTER 3

Figure 3.12: Summary of evidence for coseismic uplift and tsunami inundation for pit 43. Symbols showing the nature of diatom assemblages and upward fining sequences correspond to the schematic drawing in Figure 3.2. a) The summary of evidence for coseismic uplift highlights the increase in the ratio of freshwater to marine and brackish diatoms above sand beds as well as the increase in grass phytoliths and chrysophyte cysts above the sand beds. The microfossil content of the sands is left out to highlight the change in assemblages across the beds. b) Evidence for tsunami inundation includes the high relative abundance of tidal flat diatoms within sand beds and the presence of marine planktonic diatoms within sand beds.

96 CHAPTER 3

Figure 3.13: Summary of evidence for coseismic uplift and tsunami inundation for pit 18. Symbols showing the nature of diatom assemblages and upward fining sequences correspond to the schematic drawing in Figure 3.2. a) The summary of evidence for coseismic uplift highlights the increase in the ratio of freshwater to marine and brackish diatoms above sand beds as well as the increase in grass phytoliths and chrysophyte cysts above the sand beds. The microfossil content of the sands is left out to highlight the in assemblages across the beds. b) Evidence for tsunami inundation includes the high relative abundance of tidal flat diatoms within sand beds and the presence of marine planktonic diatoms within sand beds.

97 CHAPTER 3

TABLE 3.1. RADIOCARBON AGE DETERMINATIONS, QUINTERO, CHILE

Radiocarbon Maximum or Elevation Laboratory Lab reported Calibrated age minimum Sample #1 Pit2 MTL (m) Sand number age (14C yr BP) (cal yr BP - 2σ) limiting age3 Description of dated material Charcoal fragment immediately below 1 43 3.3 A 302252 3660 ± 40 3830-4085 maximum Sand bed A Rhizome sheath immediately below 2 43 3.3 A 302253 3450 ± 30 3567-3821 maximum Sand bed A Rhizome sheath immediately below 3 43 2.98 B 302251 3600 ± 40 3711-3974 maximum Sand bed B Rhizome sheath immediately below 4 43 2.87 C 302250 3980 ± 40 4242-4519 maximum Sand bed C Rhizome fragment immediately below 5a 18 1.88 C 295081 4020 ± 30 4297-4529 maximum Sand bed C 5b 18 1.88 C 293371 3950 ± 40 4160-4509 maximum Twig immediately below Sand bed C 5a,b pooled 18 n/a C n/a n/a 4291-4517 maximum n/a mean Rhizome sheath 0.07 m above Sand 6 43 2.62 D 302254 4440 ± 30 4857-5259 minimum bed D 7 41 2.37 E 302248 4940 ± 40 5492-5732 maximum Twigs immediately below Sand bed E Charcoal fragments immediately above 8 43 2.33 F 302249 5310 ± 40 5931-6182 minimum Sand bed F Basal Twigs in the top 0.03 m of the basal 9 17 1.47 sand 293373 5060 ± 40 5655-5896 maximum sand Twig or stem below thin, discontinuous 10 40 3.85 n/a 306347 260 ± 30 148-321 maximum sand 11 47 2.35 n/a 306351 1740 ± 30 1539-1701 maximum Charcoal within white precipitate layer Charcoal above thin, discontinuous 12 47 2.3 n/a 306352 2010 ± 30 1839-2001 minimum sand Small gastropod above a mixed marl 13 44 2.45 n/a 306348 3620 ± 30 3726-3980 minimum and sand unit Small gastropod above discontinuous 14 53 0.55 n/a 306354 200 ± 30 0-292 minimum sand layer

1 Sample ID’s and corresponding sampling locations can be found on Figure 3.6a,b. 2 Pits with corresponding radiocarbon age locations can be found on Fig. 3.5b. 3 Interpretation of the stratigraphic context of the dated sample relative to the time that host unit was deposited. Maximum ages are on samples containing carbon judged to be older than a sand bed. Minimum ages are on samples judged younger than a sand bed.

98 CHAPTER 3

TABLE 3.2. ECOLOGY OF DIAGNOSTIC DIATOM SPECIES

Species (> 5% abundance) Ecology 1 Ecology 2 Ecology 3 Life form4

Achnanthes brevipes M & B B & M M & B Epiphytic

Achnanthes brevipes var. intermedia M & B B & M M & B Epiphytic

Planothidium delicatulum B B B & M Epipsammic

Amphora coffeaeformis M & B B ---- Epipelic

Amphora ovalis FW FW FW Epipelic

Cocconeis scutellum M & B M & B M & B Epiphytic

Cocconeis diminuta B FW & B B & FW Epiphytic

Cyclotella meneghiniana B B B Tychoplanktonic

Cymbella pusilla FW FW & B ------

Cymbella silesiaca FW FW & B ---- Epipelic

Denticula elegans FW FW & B ---- Aerophilic & Epiphytic

Diploneis smithii B B & M B & M Epipelic

Diploneis subovalis FW FW & B ---- Epipelic

Epithemia adnata FW FW & B FW Epiphytic

Fallacia pygmaea M & B M & B ---- Epipelic

Staurosirella pinnata FW FW & B FW & B Planktonic & Epipelic

Gomphonema parvulum FW FW & B FW Aerophilic & Epiphytic

Gyrosigma nodiferum B B & FW B Epipelic

Navicula cryptotenella FW FW & B ---- Epipelic

Navicula libonensis FW FW & B ---- Epipelic

Navicula reinhardti FW FW & B ---- Epipelic

Navicula rhynchocephala FW FW & B B Epipelic

Navicula tripunctata B FW & B ---- Epipelic & Epiphytic

Nitzschia amphibia FW FW & B FW Aerophilic & Epiphytic

Nitzschia frustulum FW FW & B ---- Aerophilic & Epiphytic

Nitzschia lanceolata B B ---- Epipelic & Epiphytic

Nitzschia linearis FW FW ---- Epipelic

Paralia sulcata M & B M & B M Planktonic

Pinnularia brevicostata FW FW ------

Pinnularia viridis FW FW FW Aerophilic

Rhopalodia gibba FW FW FW & B Epiphytic

Rhopalodia gibberula FW ------Epiphytic

Surirella striatula M & B B & M ---- Epipelic

Tabularia fasciculata M & B B & M M & B Epiphytic

Thalassiosira lacustris M & B M & B M Planktonic

Thalassiosira weissflogii M & B B & M M Planktonic

1 From Krammer and Lange-Bertalot (1986, 1988, 1991a,b). Marine = M; Brackish = B; Freshwater = FW. 2 From Denys (1991) 3,4 From Vos and de Wolf (1988, 1993). Planktonic = diatoms that float freely in the water column and do not live attached to any substrate; Tychoplanktonic = diatoms that live in the benthos, but are commonly found in the plankton; Epipelic = diatoms that live on or just below the surface of wet muddy sediments; Epiphytic = diatoms that are attached to larger plants or other surfaces; Aerophilic = diatoms that are able to survive subaerial, temporarily dry conditions. 99

CHAPTER 4. LATE HOLOCENE TSUNAMIS AND SEISMIC LAND- LEVEL CHANGE IN SOUTH-CENTRAL CHILE *

*To be submitted to the Geological Society of America Bulletin as: Dura, T., Horton, B.P., Hong, I., Cisternas, M., Ely, L.L., Nelson, A.R., Nikitina, D., Pilarczyk, J.E., Wesson, R.L., Late Holocene tsunamis and seismic land-level change in south-central Chile.

ABSTRACT

Because records of large earthquakes on the subduction zone off south-central Chile are restricted to the historical period (~500 years), our understanding of the long-term behavior of the subduction zone is limited. Paleoseismic investigations from the tidal

Tirúa (38.3° S) and Quidico (38.1°S) rivers have provided an 1800-year record of earthquakes and tsunamis. We focused on coastal sedimentary sequences that preserved distinctive, sharp lithologic contacts that signal sudden changes in relative sea level

(RSL) related to coseismic vertical deformation of the coastline and tsunami inundation.

Results of stratigraphical, sedimentological and diatom analyses conducted on three riverbank sections confirm previous inferences of tsunami inundation and are consistent with historical accounts of coseismic land-level change during five historical earthquakes in AD 2010, 1960, 1835, 1751, and 1575. The amounts of vertical deformation inferred from diatom analysis of stratigraphy recording historical earthquakes suggest that Maule segment earthquakes result in uplift at our sites (e.g., 2010, 1835, and 1751), and Valdivia segment earthquakes result in no deformation (e.g., 1960) or subsidence (e.g., 1575). We identified four prehistoric sand beds at Tirúa, probably deposited by tsunamis, 14C dated to AD 1457-1575, 1443-1547, 256-461, and 176-336. Diatoms overlying the prehistoric tsunami deposits indicate uplift coincident with sand deposition in AD 1457-1575 and AD

256-461, which we attribute to the Maule segment. Subsidence coincident with sand deposition in AD 1443-1547, and no change in elevation in AD 176-336, is attributed to 100

CHAPTER 4 the Valdivia segment. Our results show how diatom-based reconstructions of coseismic deformation can address questions of long-term subduction-zone behavior and segmentation along the south-central Chilean subduction zone.

4.1 INTRODUCTION

The unexpected magnitudes of the Andaman-Aceh (2004; Mw 9.2) and Tohoku-oki

(2011; Mw 9.1) earthquakes and accompanying tsunamis underscored the danger of relying only on historical records of earthquakes and modern geodetic measurements when assessing seismic hazards at subduction zones (Sieh et al., 2008; Heki, 2011).

Spatially variable earthquake ruptures at subduction zones can result in an underestimation of hazards unless earthquake and tsunami records extend over many seismic cycles, hundreds of years prior to historical accounts and instrumental measurements (Moernaut et al., 2014). Stratigraphic evidence of subduction-zone earthquakes, like that discovered beneath the tidal marshes of Chile (Atwater et al., 1992;

Bartsch-Winkler and Schmoll, 1992; Cisternas et al., 2005; Nelson et al., 2009; Garrett et al., 2013; Dura et al., 2014), Alaska (Ovenshine et al., 1976; Combellick and Reger,

1994; Shennan and Hamilton, 2006), and the U.S. Pacific Northwest (Darienzo et al.,

1994; Nelson et al., 1998; Atwater and Hemphill-Haley, 1997), can be used to develop histories of earthquakes over thousands of years. Distinctive, sharp lithologic contacts in low energy intertidal environments signal sudden changes in relative sea level (RSL) caused by coseismic vertical deformation of the coastline (Plafker and Savage, 1970;

Atwater, 1987; Nelson et al., 1996; Witter et al., 2003). Widespread sand beds, deposited by tsunamis accompanying great (magnitude 8-9) earthquakes, are often found concomitant with lithologic evidence of sudden RSL change (Sawai, 2004; Kelsey et al.,

2002; Cisternas et al., 2005; Satake and Atwater, 2007; Bourgeois et al., 2009).

101 CHAPTER 4 Microfossils incorporated into coastal sediments provide an independent means of testing interpretations of earthquake-related RSL change and tsunami inundation inferred from coastal stratigraphy (Sherrod, 1999). Diatoms in particular have long been recognized as valuable RSL indicators due to their sensitivity to environmental factors including salinity, tidal exposure, and substrate, resulting in their application to earthquake and tsunami reconstructions at several subduction zones (Hemphill-Haley,

1995a; Hemphill-Haley, 1996; Shennan et al., 1996; Sawai, 2001; Sherrod, 2001;

Pilarczyk et al., 2014). However, the application of diatom biostratigraphy to earthquake and tsunami studies has mostly focused on prehistoric events, thereby making it difficult to test the efficacy of the method.

Although temporally limited, the 500-year long historical record of coseismic vertical deformation and tsunami inundation along the coasts of the south-central Chilean subduction zone does provide an opportunity to evaluate stratigraphic and diatom methods of identifying earthquakes and tsunamis, thus providing confidence to prehistoric reconstructions (Fig. 4.1). The AD 1960 and 2010 tsunamigenic earthquakes, which are known to have affected our sites based on local resident accounts, provide critical modern analogues for sedimentary and microfossil signatures of coseismic vertical deformation and tsunami inundation. This paper expands on previous studies by

Ely et al. (2014) and Hong (2014) that identified stratigraphic evidence of five historical tsunamis within marshes bordering the tidal Tirúa (38.3° S) and Quidico (38.1°S) rivers by applying diatom analysis to confirm prior mapping of tsunami inundation and identify land-level changes coincident with sand deposition. We analyzed two stratigraphic sections at Tirúa to test the intra-site variability of diatom-based RSL reconstructions and validate our interpretations of coseismic land-level change. A third

102 CHAPTER 4 stratigraphic section at Quidico—10 km north of Tirúa—allowed us to compare our diatom-based RSL reconstructions between sites and correlate events based on the nature of coseismic deformation as well as on radiocarbon age. Our detailed diatom analyses identify variable land-level change consistent with historical accounts, surveys, and models that indicate either coseismic uplift or subsidence at or near our sites depending on the slip distribution of the rupture (Moreno et al., 2012; Briggs et al.,

2014). In addition, new radiocarbon analyses identify four prehistoric earthquakes and accompanying tsunamis at Tirúa, extending the earthquake and tsunami record there to include 8 events in the last ~1800 years. Diatom evidence of uplift or subsidence associated with the prehistoric ruptures allows us to use historical slip distributions to infer slip distributions of past events. Our results show that diatom-based RSL reconstructions of historical and prehistoric coseismic deformation are a powerful tool in addressing questions of long-term subduction-zone behavior and segmentation along the south-central Chilean subduction zone.

4.2 BACKGROUND

4.2.1 Tectonic setting

The Chilean subduction zone marks the plate boundary between the subducting Nazca plate and the South American plate, extending 3500 km from the northern coast of Chile to the triple junction offshore of southern Chile (Fig. 4.1a). In south-central Chile, the

Nazca plate subducts to the northeast beneath the South American plate at about 66 mm/yr (Angermann et al., 1999). Our study sites (Tirúa and Quidico) are adjacent to the

Mocha Fracture zone (38°S), which separates the Maule segment to the north, and the

Valdivia segment to the south (Fig. 4.1b; Moreno et al., 2011). Geodetic studies show that the Maule segment and the Valdivia segment are highly coupled (Métois et al., 2012). 103 CHAPTER 4 The largest historical ruptures on the Maule segment (north of ~38°S) occurred in AD

2010, 1835, 1751, 1657, and 1570, while significant ruptures on the Valdivia segment

(south of ~38°S) occurred in AD, 1960, 1837, and 1575 (Fig. 4.1b; Montessus de Ballore,

1912; Lomnitz, 2004; Cisternas et al., 2005; Udías et al., 2012).

The geometry of the megathrust surface and the amount and distribution of coseismic and interseismic slip on that surface control the vertical displacement of coasts above subduction zones (Wang et al., 2013). A simple slip model at Tirúa indicates a deeper, landward slip zone will result in coseismic uplift, while a shallower slip zone concentrated offshore will result in coseismic subsidence (Moreno et al., 2009, 2012; Ely et al., 2014). Inferred slip distributions for the Mw 8.8 2010 and Mw 9.5 1960 earthquakes are consistent with this model; in 2010, a deeper, landward slip zone produced coseismic uplift at Tirúa, while an intermediate slip zone in 1960 produced little to no vertical change (Moreno et al., 2009 and 2012). Thus, we can infer a similar slip distribution for historical and prehistoric earthquakes displaying similar vertical deformation patterns.

4.2.2 Tirúa

The Tirúa River (drainage basin ~40 km2) meanders through a low-lying coastal floodplain vegetated by freshwater to brackish marsh species (e.g., Juncus balticus and

Spartina sp.) that change their distribution over a decimenter-scale elevational range.

The river drains into the Pacific Ocean at a northwestward facing embayment (Fig. 4.2b).

The coast is wave dominated and mesotidal (Great Diurnal tidal range of ~2.4 m between MLLW-MHHW) as measured at the closest permanent tide gauge at Lebu, 80 km north of Tirúa (Admiralty, 2004).

104 CHAPTER 4 Ely et al. (2014) used geomorphic, sedimentological, and stratigraphic methods combined with historical accounts of earthquakes and tsunamis to interpret a sequence of buried organic silts, interbedded with sand beds as evidence for repeated late

Holocene coseismic land-level change and tsunami inundation at Tirúa. Four distinct, tabular sand beds were correlated along a ~500-m-long stratigraphic transect composed of 27 stratigraphic sections along the Tirúa River bank (Fig. 4.3a). Based on OSL and

137Cs analyses and historical accounts, the two youngest sand beds were attributed to the recent earthquakes in AD 2010 and 1960. Eyewitnesses to both the 2010 and 1960 tsunamis describe sediment-laden water surging 3km up the Tirúa River, leaving a detectable deposit 2 km upriver. Radiocarbon ages and historical records indicate the two oldest sand beds were most likely deposited by tsunamigenic historical earthquakes in 1575, 1751. Ely et al. (2014) concluded that the historical tsunamis in AD 1837, 1835,

1657, and 1570 were not large enough to leave a detectable deposit 2 km up the Tirúa

River.

4.2.3 Quidico

The Quidico River (drainage basin ~20 km2) drains from Quidico Lake into the Pacific

Ocean (Fig. 4.2c). In its lower reaches, the river meanders through a low-lying coastal floodplain vegetated by freshwater to brackish marsh species (e.g., Juncus balticus,

Juncus microcephalus, Scirpus americanus, Scirpus californicus, and Spartina sp.) that change their distribution over a decimeter-scale elevational range. The mouth of the

Quidico River is protected by bedrock to the south and west, creating a northward- facing, cuspate-shaped delta. Recent construction prior to the 2010 earthquake added sea walls that extend approximately 0.5 km upstream from the mouth of the river. Like

Tirúa, which is 10 km to the south, the Quidico coast is wave dominated and mesotidal 105 CHAPTER 4 (Great Diurnal tidal range of ~2.4 m between MLLW-MHHW) as measured at Lebu

(Admiralty, 2004).

Employing similar stratigraphic methods and analysis of historical records, Hong (2014) interpreted laterally extensive sand beds at Quidico as evidence for repeated late

Holocene tsunami inundation. Five distinct, tabular sand beds were correlated along a

~300-m-long stratigraphic transect composed of 13 cores and 14 pits through an abandoned meander of the Quidico River (Fig. 4.3b). Based on 137Cs analyses and historical accounts, the two youngest sand beds were attributed to the recent earthquakes in AD 2010 and 1960. Eyewitnesses to the 2010 tsunami reported that the tsunami inundated the Quidico River floodplain where coring and pit transects are located. Based on radiocarbon ages and historical accounts, the site stratigraphy also includes tsunami deposits associated with the AD 1835 and 1751 earthquakes. A prehistoric sand bed dated to AD ~1450 was also described. Of the five sand beds preserved at Quidico, three are attributed to tsunamis accompanying earthquakes on the

Maule segment, including the AD 1835 tsunami, which is not preserved at Tirúa (Hong,

2014).

4.3 APPROACH AND METHODS

To confirm and expand on previous stratigraphic investigations by Ely et al. (2014) and

Hong (2014), we selected two representative riverbank sections at Tirúa (sections T16 and T2) and one at Quidico (section Q9) for detailed diatom analysis (Fig. 4.3a, b).

Monoliths (6 x 5 x 50 cm columns of sediment) were collected in the field. We collected section T2 from the same riverbank location used for radiocarbon and grain-size analyses in Ely et al. (2014). T2 contains the four historical tsunamis mapped by Ely et

106 CHAPTER 4 al. (2014) in 25 of 27 riverbank sections, plus one additional sand bed. The four historical tsunamis described by Ely et al. (2014) are also present in section T16, as well as four older sand beds not previously described. Section Q9 contains the four historical and one prehistoric sand bed described in 13 of 27 stratigraphic sections by Hong (2014). We conducted detailed diatom, grain-size, and organic content (loss on ignition) analyses to test inferences of tsunami inundation and identify coseismic land-level change coincident with sand deposition (Fig. 4.4a, b, 4.5). We also dated additional radiocarbon samples from section T16 to extend its earthquake and tsunami record back in time. The elevation of riverbank sections and samples were tied into benchmarks and tidal datums established by Ely et al. (2014) and Hong (2014), and reported as elevation (in meters) relative to mean tide level (MTL).

4.3.1 Diatom analysis

Subsampling intervals for diatom analysis in sections T16, T2, and Q9 ranged from 1 cm near lithologic contacts to 10-cm intervals in homogenous sediment. About 1 g of sampled sediment was oxidized with nitric acid in a microwave digestion system to remove organic material (Charles et al., 2002). Following digestion, between 25 and 100

l (depending on the diatom concentration) of liquid sample was pipetted and distributed evenly on a cover slip. The cover slip was dried overnight and then inverted and mounted on a glass slide using Naphrax (Charles et al., 2002). We prepared 70 diatom slides from sections 16 and 2 at Tirúa, and 39 from section Q9 at Quidico. We identified diatoms to species level using a light microscope under oil immersion at 1000x magnification with reference to Krammer and Lange-Bertalot (1986, 1988, 1991a,b) and

Lange-Bertalot (2000). Between 300 and 400 diatoms were identified and counted in slides, with each species expressed as a percentage of total diatom valves counted. 107 CHAPTER 4 Fragments containing more than half a valve were included in the count. Only species that exceeded 5% of total valves counted were used for paleoecological interpretation

(e.g., Horton et al., 2007).

We identified 140 species in 63 genera from 109 samples in Tirúa sections T16 and T2 and Quidico section Q9. Diatoms were classified (Table 4.1) by salinity (marine, brackish, and freshwater) and life-form (planktonic, epipelic, epiphytic, epipsammic, aerophilic) based on the ecological preferences outlined in Chilean (e.g., Rivera, 1983;

Rebolledo et al., 2005, 2011) and global (e.g., Krammer and Lange-Bertalot, 1986, 1988,

1991a,b; Denys, 1991; Vos and de Wolf, 1988, 1993; Hartely et al., 1996; Lange-Bertalot,

2000) catalogs. The marine diatom group includes marine and marine-brackish species that thrive in salinities exceeding 30 practical salinity units (psu). The brackish diatom group includes brackish-marine and brackish species that tolerate salt concentrations between 0.2-30 psu. The freshwater group includes fresh-brackish and fresh diatoms that generally occur in salt concentrations less than 0.2 psu (Vos and de Wolf, 1993).

Diatom taxa that live attached to vegetation are defined as epiphytic forms. Taxa that live on fine wet sediments are defined as epipelic forms while taxa that live attached to sand grains are defined as epipsammic forms. Taxa that live on wet sediments but are able to survive temporarily dry conditions are defined as aerophilic forms. Tychoplanktonic diatoms include an array of species that live in the benthos, but are commonly found in the plankton. Diatoms that float in the water column and do not live attached to any substrate are defined as planktonic forms (Vos and de Wolf, 1988, 1993).

We defined diatom assemblage zones using a stratigraphically constrained incremental sum-of-squares (CONISS) cluster analysis. We also applied detrended correspondence

108 CHAPTER 4 analysis (DCA) to examine the pattern of diatom assemblage variation within and between samples, and to identify environmental gradients. Higher eigenvalues represent strong environmental gradient, such as salinity, along an axis. Samples with similar species compositions are grouped together in the DCA bi-plot and samples with statistically different species compositions plot apart (e.g., Birks, 1986; 1992).

4.3.2 Grain size and loss on ignition analyses

Subsampling intervals for grain-size and loss on ignition (LOI, a proxy for the percentage of organic matter) analyses in sections T16, T2, and Q9 ranged from 1 cm near lithologic contacts to 3-cm intervals in homogenous sediment. After treating samples with hydrogen peroxide (30%) to remove organic material, samples were analyzed using a

Beckman Coulter LS230 laser diffraction grain-size analyzer. Grain-size interpretations are based on the Wentworth Phi Scale and reported as differential volume (the percentage of total volume that each size class occupies). The grain-size distributions were interpolated and gridded using a triangular irregular network (TIN) algorithm

(Sambridge et al. 1995) and plotted as color surface plots using Geosoft Oasis TM software (Donato et al. 2009). The mean grain size and dominant grain size values for the 10th (D10) percentile (diameter at which 10% of the sample volume comprises smaller particles) of each sample were used to characterize lithologic units (Table 4.2). D values take into account the skewness of a sediment distribution curve (e.g. fine tail) and are more suitable descriptors of grain size in non-Gaussian distributions (Blott and Pye,

2001; Pilarczyk et al., 2012). Lithologic descriptions of units and sand beds are based on the mean grain size value.

109 CHAPTER 4 Fossil samples for LOI were dried at 105C for 12 hours, weighed, baked at 550C for 6 hours, and weighed again to estimate the weight lost (LOI; Ball, 1964).

4.3.3 Dating

We used radiocarbon and 137Cs analyses and stratigraphic correlations from Ely et al.

(2014) and Hong (2014) to constrain the ages of the upper four sand beds in Tirúa sections T16 and T2 and the five sand beds in Quidico section Q9 (Table 4.3). For additional dating of older sand beds in section T16, plant macrofossils were collected from organic sediments that gave either limiting maximum ages (usually below each contact) or limiting minimum age (in growth position above each contact). We collected samples that were in growth position (rhizomes) or so delicate (e.g., seeds, leaf parts) that they would probably have decayed or been broken by transport soon after death, thus reducing the likelihood of dating material much older than sample burial (Kemp et al., 2013). Radiocarbon ages were calibrated using OxCal radiocarbon calibration software (Bronk Ramsey, 2009) with the SHCal13 data set of Hogg et al. (2013).

Calibrated age ranges are shown at two standard deviations in Years BC/AD. We use maximum and minimum limiting radiocarbon ages to calculate age probability distributions for inferred prehistoric tsunami beds from Tirúa section T16 with the V- sequence feature of OxCal (Table 4.4; Bronk Ramsey 2008; e.g., DuRoss et al., 2011;

Berryman et al., 2012). To estimate inter-event times and recurrence intervals of paleoearthquakes and tsunamis we used the “difference” function in OxCal.

4.4 RESULTS

We identified nine distinct 3-cm to 15-cm thick sand beds in the Tirúa and Quidico sedimentary sequences. We labeled the beds from youngest (1) to oldest (9) (Fig. 4.4a, b,

110 CHAPTER 4 4.5). When age ranges for beds at Tirúa and Quidico overlapped, we assigned the same number to the beds (e.g., bed 1 at Tirúa and bed 1 at Quidico were deposited at the same time). All beds were distinct from their host sediments due to their sharp (< 1-3 mm) upper and lower contacts, lack of organic matter, uniform thickness, and lateral extent.

An increase in mean grain size and a decreased fine fraction—shown by a decrease in

D10 value—distinguishes all sand beds from underlying and overlying units (Table 4.2).

Tirúa sections T16 and T2—separated by ~132 m —contained beds 1, 2, 4, 5, and 6 (Fig.

4.3, 4.4a, b). Section T16 contained three additional beds (beds 7, 8, and 9). The beds are interbedded within four main lithostratigraphic units (units 1, 2, 3, and 4). A brown (10

YR 3/1) organic very fine sand underlies section T16 between -0.3 and 0.05 m MTL (unit

1). A brown (10 YR 3/1) organic very fine sand (unit 2) overlies unit 1 in section T16, and forms the basal sediment in section T2. Unit 2 is 0.1 m thick in section T16 and 0.3 m thick in section T2. In both sections, ~0.7 m of organic very fine sand (unit 3)—varying in color from grey (10 YR 5/1) to brown (10 YR 5/2)—overlies unit 2. The top 0.2 m (T16) and 0.13 m (T2) of the sections consists of light brown (10 YR 6/2) organic very fine sand

(unit 4). All units contain humified organic matter. The upper portion of unit 1 contains abundant identifiable herbaceous plant fragments.

Quidico section Q9 contained beds 1, 2, 3, 4, and 7 (Fig. 4.5). The beds are interbedded within three main lithostratigraphic units (units 1, 2, and 3). A grey (10YR 2/1) medium sand underlies section Q9 between -0.05 and 0.02 m MTL (unit 1). A 0.55 m thick, brown (10YR 4/2) organic very fine sand (unit 2) overlies unit 1 (Fig. 4.5). Interbedded brown (10YR 5/2) fine sand and silt layers (unit 3) overly unit 2 between 0.62 m and

111 CHAPTER 4 0.72 m MTL. The top ~0.3 m of the core is composed of unit 2 sediment. All units contain humified organic matter, but few identifiable herbaceous plant fragments.

4.4.1 Tirúa section 16 (T16)

4.4.1.1 Sand bed 9 and underlying and overlying units (T16)

Between -0.3 and -0.25 m MTL, a brown, poorly sorted organic (mean LOI 28%) very fine sand (unit 1; mean = 3.26 Φ; D10 = 5.28 Φ) underlies bed 9 (Fig. 4.4a). The marine planktonic diatoms Paralia sulcata and Thalassiosira angulata, and the marine epipsammic Opephora pacifica dominate the assemblage within the organic very fine sand (Fig. 4.6). Minor abundances of the freshwater epipelic Amphora ovalis and epiphytic Cocconeis placentula are also present. Constrained cluster analysis classified samples underlying bed 9 into zone T16-1.

At -0.25 m MTL, a sharp contact separates unit 1 from bed 9, a 5-cm-thick, grey, well sorted, fine quartz sand (mean = 2.61 Φ; D10 = 4.51 Φ) containing little organic material

(mean LOI 7%). The bed contains a mixed diatom assemblage composed of freshwater epipelic (e.g. A. ovalis, Karayevia oblongella), brackish epipsammic (e.g. Planothidium delicatulum), and marine planktonic (e.g., P. sulcata, T. angulata) and epipsammic

(e.g., O. pacifica) taxa.

A brown, poorly sorted organic (mean LOI 26%) very fine sand (unit 1; mean = 3.01 Φ;

D10 = 5.96 Φ) overlies bed 9 at -0.2 m MTL. Diatom assemblages within the organic very fine sand are similar to assemblages underlying bed 9; the marine planktonic diatoms P. sulcata, and T. angulata and the marine epipsammic diatom O. pacifica dominate the assemblage, which also includes smaller abundances of the freshwater epipelic A. ovalis

112 CHAPTER 4 and K. oblongella. Constrained cluster analysis classified samples overlying bed 9 into zone T16-1. Zone T16-1 samples above and below bed 9 group together in the DCA biplot, reflecting the similarity in their diatom composition (Fig. 4.7a).

Leaf fragments collected from the organic very fine sand immediately underlying and overlying bed 9 yield a maximum limiting age of AD 129-314 and a minimum limiting age of AD 247-381 for sand deposition (Table 4.3; Fig. 4.4a).

4.4.1.2 Sand bed 8 and underlying and overlying units (T16)

At -0.05 m MTL, a sharp contact separates unit 1 from bed 8, a 5-cm-thick, grey, well sorted, fine quartz sand (mean = 2.40 Φ; D10 = 4.43 Φ; Fig. 4.4a) containing little organic material (mean LOI 3%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g., O. pacifica), epiphytic (e.g., Achnanthes brevipes), and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish epipsammic (e.g., Cocconeis neodiminuta) and freshwater epipelic (e.g., A. ovalis, K. oblongella) species (Fig. 4.6).

A brown, poorly sorted organic (mean LOI 25%) very fine sand (unit 1; mean = 3.05 Φ;

D10 = 5.73 Φ) overlies bed 8 at 0.0 m MTL. Within the organic very fine sand, there is an increase in freshwater epipelic (e.g., A. ovalis, Diploneis psuedovalis, K. oblongella) and epiphytic (e.g., Gomphonema parvulum, Nitzschia frustulum), and brackish epipelic

(e.g., Navicula cincta, Navicula peregrina) and epipsammic (e.g., Planothidium delicatulum) taxa compared to the marine dominated assemblage underlying bed 8. The percentage of marine and brackish planktonic diatoms decreases to 10% of the assemblage compared to 25% in sediments underlying bed 8 (Fig. 4.8). Constrained cluster analysis classified samples overlying bed 8 into zone T16-2. Zone T16-2 samples 113 CHAPTER 4 plot separately from samples underlying bed 8 (zone T16-1) in the DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7a).

Scirpus sp. seeds collected from the organic very fine sand immediately underlying and overlying bed 8 yield a maximum limiting age of AD 240-356 and a minimum limiting age of AD 417-530 for sand deposition (Table 4.3; Fig. 4.4a).

4.4.1.3 Sand bed 7 and underlying and overlying units (T16)

At 0.05 m MTL, a sharp contact separates unit 1 from bed 7, a 15-cm-thick, grey, well sorted, fine quartz sand (mean = 2.17 Φ; D10 = 3.44 Φ; Fig. 4.4a) containing little organic material (mean LOI 3%). Marine planktonic (e.g., P. sulcata, T. angulata) and marine epipsammic (e.g., O. pacifica) taxa dominate the assemblage in the bed (Fig.

4.6). Minor abundances of the brackish epipelic Diploneis smithii and epipsammic C. neodiminuta and the freshwater epipelic A. ovalis are also present.

A brown, poorly sorted very fine sand (unit 2; mean = 3.48 Φ; D10 = 7.18 Φ) containing little organic material (10%) overlies bed 7 at 0.2 m MTL. Within the very fine sand, there is an increase in marine planktonic (e.g., P. sulcata, T. angulata) and epipsammic

(e.g., O. pacifica), and brackish epipelic (e.g., Amphora coffeaeformis, D. smithii,

Rhopalodia brebissonii) taxa compared to the freshwater and brackish dominated assemblage underlying bed 7. Constrained cluster analysis classified samples overlying bed 7 into zone T16-3. Zone T16-3 samples plot separately from samples underlying bed

7 (zone T16-2) in the DCA biplot, reflecting the dissimilarity in their species composition

(Fig. 4.7a).

114 CHAPTER 4 An in situ rhizome collected from the organic very fine sand below bed 7 yields a maximum limiting age for sand deposition of AD 1440-1500 (Table 4.3; Fig. 4.4a).

4.4.1.4 Sand bed 6 and underlying and overlying units (T16)

At 0.3 m MTL, a sharp contact separates unit 2 from bed 6, a 2-cm-thick, light grey, well sorted, fine quartz sand (mean = 2.18 Φ; D10 = 3.11 Φ; Fig. 4.4a) containing little organic material (mean LOI 3%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish (e.g., Amphora coffeaeformis, Navicula cincta) and freshwater (e.g.,

Navicula cari) epipelic species (Fig. 4.6).

A dark grey, poorly sorted organic (mean LOI 16%) very fine sand (unit 3; mean = 3.64

Φ; D10 = 7.81 Φ) overlies bed 6 at 0.32 m MTL. Within the organic very fine sand, there is an increase in freshwater epipelic (e.g., D. psuedovalis, K. oblongella, Nitzschia frustulum, Pinnularia intermedia, Pinnularia brevicostata) and epiphytic (e.g.,

Cosmioneis pusilla) taxa compared to the marine and brackish dominated assemblage underlying bed 6. Immediately above bed 6, the percentage of marine and brackish planktonic diatoms decreases to 11% of the assemblage compared to 20% in sediments underlying bed 6 (Fig. 4.8). Constrained cluster analysis classified samples overlying bed 6 into zone T16-4. Zone T16-4 samples plot separately from samples underlying bed

6 (zone T16-3) in the DCA biplot, reflecting the dissimilarity in their species composition

(Fig. 4.7a).

Seed casings collected from the organic very fine sand immediately above bed 6 yield a minimum limiting age of AD 1451-1619 for sand deposition (Table 4.3; Fig. 4.4a).

115 CHAPTER 4 4.4.1.5 Sand bed 5 and underlying and overlying units (T16)

At 0.65 m MTL, a sharp contact separates unit 3 from bed 5, a 3-cm-thick, grey, well sorted, fine quartz sand (mean = 2.35 Φ; D10 = 3.36 Φ; Fig. 4.4a) containing little organic material (mean LOI 5%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish (e.g., Amphora coffeaeformis, Navicula cincta) and freshwater (e.g., K. oblongella, Navicula cari, Neidium alpinum, Nitzschia frustulum) epipelic species (Fig. 4.6).

A brown, poorly sorted organic (15%) very fine sand (unit 3; mean = 3.49 Φ; D10 = 6.80

Φ) overlies bed 5 at 0.68 m MTL. Within the organic very fine sand, there is a slight increase in marine planktonic (e.g., P. sulcata, T. angulata) and epipsammic (e.g., O. pacifica), and brackish epipelic (e.g., A. coffeaeformis, Tryblionella levidensis) taxa compared to the freshwater dominated assemblage underlying bed 5 (Fig. 4.6).

Constrained cluster analysis classified samples overlying bed 5 into zone T16-5. Zone

T16-5 samples plot separately from zone T16-4 samples in the DCA biplot, although the shift in axis one score is not as pronounced as the other contacts in the section, reflecting a subtle change in species composition between the zones (Fig. 4.7a).

Radiocarbon age determinations for bed 5 come Ely et al. (2014 and are described in our section T2.

4.4.1.6 Sand bed 4 and underlying and overlying units (T16)

At 0.89 m MTL, a sharp contact separates unit 3 from bed 4, a 4-cm-thick, grey, well sorted, fine quartz sand (mean = 2.33 Φ; D10 = 3.76 Φ; Fig. 4.4a) containing little

116 CHAPTER 4 organic material (mean LOI 3%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata) taxa, as well as brackish epipsammic (e.g., Planothidium delicatulum) and epipelic (e.g.,

Tryblionella levidensis), and freshwater epipelic (e.g., D. psuedovalis, Hantzschia amphioxys) and epiphytic (e.g., C. pusilla) species (Fig. 4.6).

A dark brown, poorly sorted organic (mean LOI 20%) very fine sand (unit 3; mean =

3.80 Φ; D10 = 7.58 Φ) overlies bed 4 at 0.93 m MTL. Within the organic very fine sand, there is a significant increase in freshwater epiphytic (e.g., C. pusilla) and epipelic (e.g.,

Luticola mutica, H. amphioxys, Pinnularia intermedia, Nitzschia Frustulum) taxa compared to the brackish and marine dominated assemblage underlying bed 4.

Immediately above bed 4, the percentage of marine and brackish planktonic diatoms decreases to 5% of the assemblages compared to 20% in sediments underlying bed 4

(Fig. 4.8). Constrained cluster analysis classified samples overlying bed 4 into zone T16-

6. Zone T16-6 samples plot separately from samples underlying bed 4 (zone T16-5) in the

DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7a).

Radiocarbon age determinations for bed 4 come Ely et al. (2014 and are described in our section T2.

4.4.1.7 Sand bed 2 and underlying and overlying units (T16)

Five centimeters below bed 2, the organic content of the very fine sand decreases (unit

4), freshwater diatoms slightly decrease in abundance and the brackish epipelic

Tryblionella levidensis increases in abundance (Fig. 4.4a, 4.6). Constrained cluster analysis classified samples immediately underlying bed 2 into zone T16-7 (Fig. 4.6).

117 CHAPTER 4 At 1.02 m MTL, a sharp contact separates unit 4 from bed 2, a 4-cm-thick, light grey, well sorted, fine quartz sand (mean = 2.31 Φ; D10 = 3.41 Φ) containing little organic material

(mean LOI 3%). Marine planktonic (e.g., P. sulcata, Rhaphoneis amphiceros, T. angulata, Thalassiosira pacifica, Thalassiosira tenera) and marine epipsammic (e.g., O. pacifica) taxa dominate the assemblage in the bed.

A light brown, poorly sorted organic (mean LOI 20%) very fine sand (unit 4; mean = 3.56

Φ; D10 = 7.04 Φ) overlies bed 2 at 1.06 m MTL. Diatom assemblages within the organic very fine sand are similar to assemblages underlying bed 2; the freshwater epipelic H. amphioxys and Nitzschia Frustulum and brackish epipelic Tryblionella levidensis dominate the assemblage. Constrained cluster analysis classified samples overlying bed 2 into zone T16-7. Zone T16-7 samples above and below bed 2 group together in the DCA biplot, reflecting the similarity in their diatom composition (Fig. 4.7a).

The timing of deposition for bed 2 was determined by 137Cs analysis in Ely et al. (2014), and is described in section T2.

4.4.1.8 Sand bed 1 (T16)

At 1.18 m MTL, a sharp contact separates unit 4 from bed 1, a 2-cm-thick, grey, well sorted, fine quartz sand (mean = 2.04 Φ; D10 = 2.84 Φ; Fig. 4.4a) containing little organic material (mean LOI 3%). Marine planktonic (e.g., P. sulcata, T. angulata, T. pacifica, T. tenera) and marine epipsammic (e.g., O. pacifica) taxa dominate the assemblage in the bed (Fig. 4.6).

Bed 1 was deposited by the tsunami that struck Tirúa on 27 February 2010 (Ely et al.,

2014).

118 CHAPTER 4 4.4.2 Tirúa section 2 (T2)

4.4.2.1 Sand bed 6 and underlying and overlying units (T2)

Between -0.18 and 0.11 m MTL, a brown, poorly sorted organic (mean LOI 40%) very fine sand (unit 2; mean = 3.95 Φ; D10 = 7.19 Φ) underlies bed 6 (Fig. 4.4b). The brackish epipelic A. coffeaeformis, D. smithii, and R. brebissonii, and the marine planktonic diatoms P. sulcata and T. angulata, dominate the assemblage within the organic very fine sand (Fig. 4.9). The freshwater epipelic diatom K. oblongella is also present in minor abundances. Constrained cluster analysis classified samples underlying bed 6 into zone T2-1.

At 0.11 m MTL, a sharp contact separates unit 2 from bed 6, a 3-cm-thick, grey, well sorted, fine quartz sand (mean = 2.09 Φ; D10 = 2.96 Φ) containing little organic material

(mean LOI 7%). The bed contains a mixed diatom assemblage composed of marine planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish (e.g., A. coffeaeformis,

Navicula cincta) and freshwater (e.g., K. oblongella, Navicula cari, P. intermedia) epipelic species.

A brown, poorly sorted organic (mean LOI 26%) very fine sand (unit 3; mean = 3.12 Φ;

D10 = 5.91 Φ) overlies bed 6 at 0.14 m MTL. Within the organic very fine sand, there is an increase in freshwater epipelic (e.g., G. Parvulum, K. oblongella, Nitzschia

Frustulum, Pinnularia intermedia, Pinnularia brevicostata) taxa compared to the brackish and marine dominated assemblage underlying bed 6. Immediately above bed 6, the percentage of marine and brackish planktonic diatoms decreases to 5% of the assemblage compared to 23% in sediments underlying bed 6 (Fig. 4.10). Constrained cluster analysis classified samples overlying bed 6 into zone T2-2. Zone T2-2 samples 119 CHAPTER 4 plot separately from samples underlying bed 6 (zone T2-1) in the DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7b).

The timing of deposition for bed 6 was determined by stratigraphical correlations with bed 6 in section T16.

4.4.2.2 Sand bed 5 and underlying and overlying units (T2)

At 0.42 m MTL, a sharp contact separates unit 3 from bed 5, a 5-cm-thick, grey, well sorted, fine quartz sand (mean = 2.29 Φ; D10 = 3.40 Φ; Fig. 4.4b) containing little organic material (mean LOI 4%). The bed contains a mixed diatom assemblage composed mostly of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as minor amounts of brackish (e.g., Amphora coffeaeformis) and freshwater (e.g., D. psuedovalis, Navicula cari, Neidium alpinum,

Nitzschia frustulum) epipelic species (Fig. 4.9).

A brown, poorly sorted organic (28%) very fine sand (unit 3; mean = 3.64 Φ; D10 = 7.35

Φ) overlies bed 5 at 0.47 m MTL. Within the organic very fine sand, there is a significant increase in the marine epiphyte Achnanthes brevipes and the brackish epipelons D. smithii, Fragilaria subsalina, and Navicula cincta, and a slight increase in marine planktonic (e.g., P. sulcata) and epipsammic (e.g., O. pacifica) taxa compared to the freshwater dominated assemblage underlying bed 5. Constrained cluster analysis classified samples overlying bed 5 into zone T2-3. Zone T2-3 samples plot separately from samples underlying bed 5 (zone T2-2) in the DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7b).

120 CHAPTER 4 A Rhizome sheath and a Calystegia seed collected from the organic very fine sand underlying bed 5 yield maximum limiting ages of AD 1410-1450 and 1500-1630, respectively. A wood fragment collected from the organic very fine sand above bed 5 yields a minimum limiting age of AD 1440-1630 for sand deposition (Table 4.3; Fig.

4.4b).

4.4.2.3 Sand bed 4 and underlying and overlying units (T2)

At 0.64 m MTL, a sharp contact separates unit 3 from bed 4, a 9-cm-thick, grey, well sorted, fine quartz sand (mean = 2.21Φ; D10 = 2.96 Φ; Fig. 4.4b) containing little organic material (mean LOI 4%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata and T. angulata) taxa, as well as brackish epipsammic (e.g., Planothidium delicatulum) and epipelic (e.g.,

Tryblionella levidensis), and freshwater epipelic (e.g., D. psuedovalis, H. amphioxys, L. mutica) and epiphytic (e.g., C. pusilla) species (Fig. 4.9).

A brown, poorly sorted organic (mean LOI 35%) very fine sand (unit 3; mean = 3.71 Φ;

D10 = 6.94 Φ) overlies bed 4 at 0.73 m MTL. Within the organic very fine sand, there is a significant increase in freshwater epiphytic (e.g., C. pusilla) and epipelic (e.g., D. psuedovalis, Luticola mutica, H. amphioxys, Nitzschia Frustulum) taxa compared to the marine and brackish dominated assemblage underlying bed 4. Immediately above bed 4, the percentage of marine and brackish planktonic diatoms decreases to 4% of the assemblage compared to 31% in sediments underlying bed 4 (Fig. 4.10). Constrained cluster analysis classified samples overlying bed 4 into zone T2-4. Zone T2-4 samples plot separately from samples underlying bed 4 (zone T2-3) in the DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7b).

121 CHAPTER 4 Charcoal fragments collected from the organic very fine sand immediately underlying and overlying bed 4 yield a maximum limiting age of AD 1460-1640 and a minimum limiting age of AD 1644-1800 for sand deposition (Table 4.3; Fig. 4.4b).

4.4.2.4 Sand bed 2 and underlying and overlying units (T2)

Ten centimeters below bed 2, brackish epipelic (e.g., Navicula cincta, Tryblionella levidensis, R. brebissonii) and marine planktonic (e.g., P. sulcata, T. angulata) taxa increase in abundance (Fig. 4.9). Constrained cluster analysis classified samples immediately underlying bed 2 into zone T2-5.

At 0.92 m MTL, a sharp contact separates unit 3 from bed 2, a 3-cm-thick, light grey, well sorted, fine quartz sand (mean = 2.23 Φ; D10 = 3.55 Φ; Fig. 4.4b) containing little organic material (mean LOI 3%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., Delphineis kippae, P. sulcata, T. angulata) taxa, as well as brackish epipelic (e.g., A. coffeaeformis), and freshwater epipelic (e.g., H. amphioxys, L. mutica) and epiphytic (e.g., C. pusilla) species.

A light brown, poorly sorted organic (mean LOI 25%) very fine sand (unit 4; mean = 3.33

Φ; D10 = 7.09 Φ) overlies bed 2 at 0.95 m MTL. Diatom assemblages within the organic very fine sand are similar to assemblages underlying bed 2. Brackish epipelic (e.g.,

Navicula cincta, Tryblionella levidensis, R. brebissonii), marine planktonic (e.g., P. sulcata, T. angulata), and freshwater epipelic (e.g. D. psuedovalis, H. amphioxys) taxa dominate the assemblage. Constrained cluster analysis classified samples overlying bed 2 into zone T2-5. Zone T2-5 samples above and below bed 2 group together in the DCA biplot, reflecting the similarity in their diatom composition (Fig. 4.7b). 122 CHAPTER 4 Non-natural concentrations of 137Cs were measured in sediment samples above and closely below bed 2, but not lower in the section. The absence of 137Cs lower in the section provides a maximum age of AD ~1950, a decade before high concentrations of 137Cs were introduced into the atmosphere by nuclear testing (Ely et al., 1992).

4.4.2.5 Sand bed 1 (T2)

At 1.09 m MTL, a sharp contact separates unit 4 from bed 1, a 2-cm-thick, grey, well sorted, fine quartz sand (mean = 2.14 Φ; D10 = 4.10 Φ; Fig. 4.4b) containing little organic material (mean LOI 4%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., P. sulcata, and

T. angulata) taxa, as well as brackish (e.g., A. coffeaeformis, Navicula cincta,

Tryblionella levidensis) and freshwater (e.g., H. amphioxys, L. mutica, Nitzschia

Frustulum) epipelic species (Fig. 4.9).

Bed 1 was deposited by the AD 2010 tsunami that struck Tirúa (Ely et al., 2014).

4.4.3 Quidico section 9 (QU9)

4.4.3.1 Sand bed 7 and underlying and overlying units (QU9)

Between 0.25 and 0.35 m MTL, a brown, poorly sorted organic (mean LOI 50%) very fine sand (unit 2; mean = 3.17 Φ; D10 = 5.43 Φ) underlies bed 7 (Fig. 4.5). Freshwater epipelic taxa (e.g., C. pusilla, Denticula elegans, Diploneis psuedovalis, Eunotia praerupta, L. mutica, Nitzschia frustulum, P. intermedia) make up the diatom assemblage in the organic very fine sand (Fig. 4.11). Constrained cluster analysis classified samples underlying bed 7 into zone Q9-1. Zone Q9-1 samples group together in the DCA biplot, reflecting the similarity in their diatom composition (Fig. 4.7c).

123 CHAPTER 4 At 0.35 m MTL, a sharp contact separates unit 2 from bed 7, an 11-cm-thick, grey, well sorted, fine quartz sand (mean = 2.18 Φ; D10 = 3.00 Φ; Fig. 4.5) containing 18% organic material. The bed contains a mixed diatom assemblage composed of marine epipsammic

(e.g. O. pacifica) and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish epipelic (e.g., A. coffeaeformis, Navicula cincta, R. brebissonii), and freshwater epipelic

(e.g., H. amphioxys, F. brevistriata) and epiphytic (e.g., C. pusilla) species.

A brown, poorly sorted organic (mean LOI 60%) very fine sand (unit 2; mean = 3.29 Φ;

D10 = 7.33 Φ) overlies bed 7 at 0.46 m MTL. Within the organic very fine sand, there is an increase in brackish epipelic (e.g., A. coffeaeformis, D. smithii, R. brebissonii) taxa compared to the freshwater and brackish dominated assemblage underlying bed 7.

Constrained cluster analysis classified samples overlying bed 7 into zone Q9-2. Zone Q9-

2 samples plot separately from samples underlying bed 7 (zone Q9-1) in the DCA biplot, reflecting the dissimilarity in their species composition.

Potamogeten sp. seeds and Scirpus sp. seeds collected from the organic very fine sand immediately underlying and overlying bed 7 yield maximum and limiting ages for sand deposition of AD 1385-1450 and AD 1430-1500, respectively (Table 4.3; Fig. 4.5).

4.4.3.2 Sand bed 4 and underlying and overlying units (QU9)

At 0.58 m MTL, a sharp contact separates unit 2 from bed 4, a 3-cm-thick, grey, well sorted, fine quartz sand (mean = 2.35 Φ; D10 = 3.14 Φ; Fig. 4.5) containing 15% organic material (Fig. 4.5). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (P. sulcata, T. angulata) taxa, as well as brackish epipelic (e.g., D. smithii, Navicula cincta, Navicula peregrina, R. brebissonii),

124 CHAPTER 4 and freshwater tychoplanktonic (e.g., F. construens, F. brevistriata) and epipelic (e.g., D. psuedovalis) species (Fig. 4.11).

Interbedded organic (mean LOI 25%) fine sand and silt layers (unit 3; mean = 2.64 Φ;

D10 = 5.47 Φ) overly bed 4 at 0.61 m MTL. Within the unit, there is an increase in freshwater epipelic (e.g., D. psuedovalis, F. construens, Nitzschia Frustulum) taxa compared to the brackish and marine dominated assemblage underlying bed 4.

Immediately above bed 4, the percentage of marine and brackish planktonic diatoms decreases to 3% of the assemblage compared to 17% in sediments underlying bed 4 (Fig.

4.12). Constrained cluster analysis classified samples overlying bed 4 into zone Q9-3.

Although samples above and below bed 4 contain a mixed assemblage, the slight increase in freshwater taxa and the decrease in marine planktonic taxa cause zone Q9-3 samples to plot separately from samples underlying bed 4 (zone Q9-2) in the DCA biplot, reflecting the dissimilarity in their species composition (Fig. 4.7c).

Scirpus sp. and Potamogeten sp. seeds collected from the organic very fine sand immediately below bed 4 yields a maximum limiting age for sand deposition of AD 1665-

1815 (Table 4.3; Fig. 4.5).

4.4.3.3 Sand bed 3 and underlying and overlying units (QU9)

Five centimeters below bed 3, the organic content of the very fine sand decreases and marine epipelic (e.g., D. bombus), epiphytic (e.g., O. pacifica), and planktonic (e.g., P. sulcata, T. angulata) taxa increase in abundance (Fig. 4.5, 4.11). Constrained cluster analysis classified samples immediately underlying bed 3 into zone Q9-4. Zone Q9-4 samples group together in the DCA biplot, reflecting the similarity in their diatom composition (Fig. 4.7c). 125 CHAPTER 4 At 0.72 m MTL, a sharp contact separates unit 2 from bed 3, a 5-cm-thick, grey, well sorted, fine quartz sand (mean = 2.29 Φ; D10 = 2.99 Φ; Fig. 4.5) containing little organic material (mean LOI 3%). The bed contains a mixed diatom assemblage composed of marine epipelic (e.g., D. bombus), epipsammic (e.g., O. pacifica), and planktonic (e.g., P. sulcata, T. angulata) taxa, as well as brackish epipsammic (e.g., Planothidium delicatulum) and epipelic (e.g., Navicula cincta, Tryblionella levidensis), and freshwater epipelic (e.g., H. amphioxys, K. oblongella, L. mutica, Nitzschia frustulum) species.

A brown, poorly sorted organic (mean LOI 23%) very fine sand (unit 2; mean = 3.24 Φ;

D10 = 7.59 Φ) overlies bed 3 at 0.77 m MTL. Within the organic very fine sand, there is an increase in freshwater epipelic (e.g., Luticola mutica, Nitzschia Frustulum,

Pinnularia intermedia) and tychoplanktonic (e.g., F. brevistriata) taxa compared to the marine and brackish dominated assemblage underlying bed 3. Immediately above bed 3, the percentage of marine and brackish planktonic diatoms decreases to 6% of the assemblage compared to 20% in sediments underlying bed 3 (Fig. 4.12). Constrained cluster analysis classified samples overlying bed 3 into zone Q9-5. Zone Q9-5 samples plot separately from samples underlying bed 3 (zone Q9-4) in the DCA biplot, reflecting the dissimilarity in their species composition.

Scirpus sp. seeds collected from the unit immediately below bed 3 yield a modern radiocarbon age (post-17th century) for sand deposition (Table 4.3; Fig. 4.5).

4.4.3.4 Sand bed 2 and underlying and overlying units (QU9)

Ten centimeters below bed 2, the organic content of the very fine sand decreases and marine planktonic (e.g., P. sulcata, T. angulata) and brackish epipelic (e.g., A.

126 CHAPTER 4 coffeaeformis) taxa increase in abundance (Fig. 4.5, 4.11). Constrained cluster analysis classified samples immediately underlying bed 2 into zone Q9-6.

At 0.95 m MTL, a sharp contact separates unit 2 from bed 2, a 3-cm-thick, light grey, well sorted, fine quartz sand (mean = 2.26 Φ; D10 = 3.02 Φ; Fig. 4.5) containing little organic material (mean LOI 6%). The bed contains a mixed diatom assemblage composed of marine epipsammic (e.g. O. pacifica) and planktonic (e.g., Delphineis kippae, P. sulcata T. angulata) taxa, as well as brackish epipelic (e.g., A. coffeaeformis), and freshwater epipelic (e.g., H. amphioxys, L. mutica) and epiphytic (e.g., C. pusilla) species.

A brown, poorly sorted organic (mean LOI 22%) very fine sand (unit 2; mean = 3.39 Φ;

D10 = 8.07 Φ) overlies bed 2 at 0.98 m MTL. Diatom assemblages within the organic very fine sand are similar to assemblages underlying bed 2. Marine planktonic (e.g., P. sulcata, T. angulata) and epipsammic (e.g., O. pacifica), and brackish epipelic (e.g., A. coffeaeformis) taxa dominate the assemblage. Constrained cluster analysis classified samples overlying bed 2 into zone Q9-6. Zone Q9-6 samples above and below bed 2 group together in the DCA biplot, reflecting the similarity in their diatom composition

(Fig. 4.7c).

Non-natural concentrations of 137Cs were measured in sediment samples above and closely below bed 2, but not lower in the section. The absence of 137Cs lower in the section provides a maximum age of AD ~1950, a decade before high concentrations of 137Cs were introduced into the atmosphere by nuclear testing (Fig. 4.5; Ely et al., 1992).

Bed 1 was mapped in the Quidico study area, but was not recovered in section Q9. Bed 1 was deposited by the AD 2010 tsunami (Hong, 2014). 127 CHAPTER 4 4.5 DISCUSSION

We use diatom biostratigraphy, lithology (grain-size distribution and LOI), radiocarbon analyses, and stratigraphic correlations to infer the depositional environment represented by each lithologic unit and the timing of environmental changes at Tirúa and

Quidico. The lithofacies in our most complete stratigraphic section (section T16) record an ~1800-yr-long paleoenvironmental history of the Tirúa River lowland. Prior to the deposition of beds 9 (AD 176-336) and 8 (AD 256-461), tidal flat and marine planktonic and littoral diatoms were at their highest abundance within the very fine sand of unit 1, indicating an increased tidal influence in the marsh bordering the Tirúa River (Fig. 4.6,

4.8). Unit 1 forms a prominent bench along the riverbank that protrudes from the overlying unit 2 sediments. Radiocarbon ages reveal a probable unconformity between the top of the bench (unit 1 sediments above bed 8) and bed 7, which are separated by

~1000 years but by only 8 cm of sediment (Fig. 4.4a). By AD ~1400, diatoms in sections

T16, T2, and Q9 are composed of a mixed freshwater, brackish, and marine assemblage typical of a low marsh environment, indicating a general freshening trend over the

~1200 years represented by sediments at the base of section T16 (Fig 4.6, 4.9). Over the last ~600 years, diatoms indicate minimal net RSL change. However, assemblages fluctuate back and forth suddenly between marine and/or brackish dominated and brackish and/or freshwater dominated environments above or below the anomalous, laterally extensive sand beds, a pattern we attribute to coastal vertical deformation associated with large tsunamigenic earthquakes.

The relatively complete record of earthquakes and tsunamis at Tirúa and Quidico over the last ~600 years is likely the result of relatively stable RSL over that time (e.g., Garrett et al., 2014), while the fragmentary record found at the base of T16 is typical of falling

128 CHAPTER 4 RSL and reworking of sediments. The fragmentary nature of paleoseismic records at sites recording mid-Holocene RSL highstands (e.g., Dura et al., 2011, 2014; Brill et al.,

2013) versus the more complete records found on coasts with rising or stable Holocene

RSL (e.g. Kelsey et al., 2002; Witter et al., 2003), where accommodation space is available for preservation of earthquakes and tsunami evidence, highlights the importance of RSL on the preservation of coastal stratigraphic evidence of earthquakes and tsunamis.

4.5.1 Evidence for tsunami deposition

Ely et al. (2014) and Hong (2014) inferred a tsunami origin for the sand beds within the

Tirúa (beds 1, 2, 4, and 5) and Quidico (beds 1, 2, 3, and 4) sequences based on characteristics shared by all beds including: significant lateral extent (> 200 m), uniform thickness, and sharp (1-3 mm) lower and upper contacts (Fig. 4.3a, b). Such characteristics have been found in historical tsunami deposits from Papua New Guinea

(e.g., Gelfenbaum and Jaffe, 2003; Dawson, 2007), Thailand (e.g., Jankaew et al., 2008;

Sawai et al., 2009), Sumatra (e.g., Monecke et al., 2008), Chile (e.g., Cisternas et al.,

2005; Horton et al., 2011; Garrett et al., 2013; Ely et al., 2014), and Japan (e.g., Goto et al., 2011; Szczuciński et al., 2012). In addition, witnesses at Tirúa and Quidico describe tsunami surges flooding our study sites in AD 1960 and 2010, supporting a tsunami origin for the two youngest sand beds (beds 1 and 2), and providing analogues for older events. Our results support a tsunami origin for beds 1, 2, 3, 4, and 5 as described by Ely et al. (2014) and Hong (2014), in addition to supporting a tsunami origin for the newly described beds 6, 7, 8, and 9 based on the sedimentological and diatom characteristics they share with known historical tsunami deposits.

129 CHAPTER 4 Sand beds 3-9 in sections T16, T2, and Q9 have very similar sedimentologies to sand beds deposited by tsunamis in AD 1960 and 2010 (Table 4.2). Beds 1 and 2 are composed of fine sand (mean grain size = 2.20 Φ) and display a low fine fraction (mean D10 = 3.38

Φ) and low organic content (mean LOI = 3%; Table 4.2). Similarly, beds 3-9 are composed of fine sand (mean grain size = 2.26 Φ), and display a low fine fraction (mean

D10 = 3.45 Φ) and low organic content (mean LOI = 3%; Table 4.2; Fig. 4.4a, b, 4.5), supporting a tsunami source for the sands. Hong et al. (2014) reported that sandy beach, dune, and tidal river channel sediments at Quidico displayed a similar sedimentology to sand beds, further supporting an oceanward source for beds in our sections.

The diatom floras of the beds are typical of tsunami deposits: beds contain a mixture of freshwater, brackish, and marine diatoms (Fig. 4.8, 4.10, 4.12) due to the fact that tsunamis erode, transport, and deposit marine, brackish, and freshwater sediments (with associated taxa) as they inundate coastal and inland areas (e.g., Dawson et al., 1996;

Dawson and Smith, 2000; Smith et al., 2004; Bondevik et al., 2005; Sawai et al., 2008).

From such mixed assemblages, anomalous marine and brackish diatoms, especially offshore planktonic taxa, can support a seaward source of the sediment (e.g., Dominey-

Howes et al., 2000; Goff et al., 2012; Szczuciński et al., 2012). At the tidal Tirúa and

Quidico rivers, it is not uncommon to find marine and brackish epipsammic and planktonic species in the organic fluvial sediment below and above the beds. However, sand beds 1 and 2, deposited by known historical tsunamis in AD 1960 and 2010, display an increase in marine planktonic diatoms, providing analogues for older events (Fig. 4.8,

4.10, 4.12). Similar increases in marine planktonic diatoms are observed in beds 5 and 7, supporting an oceanward source for the sands.

130 CHAPTER 4 The lateral continuity and tabular geometry of sand beds 1-6 helps rule out localized processes, such as tidal channel migration, liquefaction, or sand venting during earthquakes as sources for sand bed deposition (Fig. 4.3; e.g., Atwater, 1992; Allen,

2000; Williams et al., 2005; Arcos et al., 2012). Sands 7-9 are not as widely mapped throughout the lowlands, however coseismic land-level change concurrent with the sand deposition supports a tsunami origin for the sands.

4.5.2 Evidence for sudden, widespread, and lasting land-level change

We infer that the unusually sudden, widespread, and lasting changes in diatom assemblages associated with the sand beds signal sudden RSL change during times of decimeter-scale coseismic deformation (Kelsey et al., 2002). In sections T16, T2, and Q9, sudden changes in diatoms underlying and overlying beds represent a record of both coseismic uplift and subsidence (Fig. 4.8, 4.10, 4.12). Our qualitative interpretations of changes (or lack of changes) in diatom species composition in sediments underlying and overlying beds are supported by cluster analysis and DCA results (Fig. 4.7; Table 4.5).

The DCA biplots show that axis one is correlated with salinity (Figure 4.7). In section

T16, the first two DCA axes of fossil assemblages account for 31% and 10% of the total variance of species data (Fig. 4.7; Table 4.5). In sections T2 and Q9, the first two DCA axes account for 37% and 10% and 25% and 10% of the total variance of species, respectively (Fig. 4.7; Table 4.5). Axis One scores for fossil assemblages imply that Axis

One reflects a salinity gradient from a freshwater to high salt-marsh environment dominated by freshwater and fresh-brackish assemblages through a low-marsh to tidal flat environment dominated by marine assemblages (Fig. 4.7; Table 4.5).

131 CHAPTER 4 At Tirúa, paleoecological changes spanning beds 1, 2, 4, 5, and 6, which were correlated over the 132 m separating sections T16 and T2, are consistent in the two sections, indicating minimal intra-site variability in diatom response to RSL change (Fig. 4.3a,

4.8, 4.9). Beds 7, 8, and 9 were only described in section T16, however diatoms overlying the beds display similar paleoecological changes to those observed above widely correlated beds. The lateral continuity of the diatom evidence for sudden RSL change coincident with beds 1, 2, 4, 5, and 6 at Tirúa indicates large portions of the marsh were affected and helps rule out gradual RSL change as a mechanism for paleoecological shifts associated with the beds (Hemphill-Haley, 1995a). Although we note minor inter-site variability between common diatom species at Tirúa and those found within Quidico section 9, we observe similar shifts in species composition between diatoms underlying and overlying beds 2, 4, and 7 at both sites (Fig. 4.8, 4.10, 4.12). The variability in diatom species between sites is expected, as modern diatom studies in Chilean marshes have encountered difficulties matching modern and fossil diatoms due to the inter-site variability of assemblages (Nelson et al., 2009; Garrett et al., 2013; Garrett et al., 2014).

An increase in freshwater diatoms in units overlying beds 3, 4, 6, and 8 (Fig. 4.8, 4.10,

4.12) is consistent with RSL fall caused by coseismic uplift. Sediment underlying the beds contains higher abundances of marine epipsammic (e.g., O. pacifica), planktonic (e.g., P. sulcata, T. angulata), and brackish epipelic (e.g., Navicula cincta, D. smithii, R. brebissonii) diatoms than overlying sediment, which displays an increase in freshwater taxa (e.g., D. psuedovalis, H. amphioxys, K. oblongella, Nitzschia frustulum, Pinnularia intermedia) and a significant decrease in marine planktonic diatoms (Fig. 4.6, 4.11, 4.9).

We also measured an increase in LOI, indicative of reduced tidal inundation and

132 CHAPTER 4 increased marsh vegetation, and hence RSL fall, in sediment overlying beds 3, 4, 6, and 8

(e.g., Nelson et al., 1996; Witter et al., 2003; Fig. 4.4a, b, 4.5).

An increase in the abundance of marine and brackish diatoms in units overlying beds 5 and 7 is consistent with RSL rise caused by coseismic subsidence. Diatom assemblages in sediment underlying the beds contain abundant freshwater taxa (e.g., D. psuedovalis,

Cavinula lapidosa, K. oblongella, Nitzschia frustulum, Pinnularia intermedia), while sediment overlying the bed displays an increase in brackish epipelic (e.g., A. coffeaeformis, D. smithii, F. subsalina) and marine epiphytic (e.g., Achnanthes brevipes), planktonic (e.g., P. sulcata, T. angulata), and epipsammic (e.g., O. pacifica) taxa. We also measured a decrease in LOI indicative of increased tidal inundation and decreased marsh vegetation, and hence RSL rise, in sediment overlying beds 5, and 7

(Shennan et al., 2009; Fig. 4.4a,b, 4.5).

The diatom assemblages in units overlying the sand beds display a lasting shift in paleoecology, rather than a temporary response to short-term sea-level fluctuations (e.g.,

El Niño events, storms; Witter et al., 2001) or the river flooding the lowland (Hemphill-

Haley, 1995a). Above bed 4 at Tirúa, and beds 3 and 4 at Quidico, the influx of freshwater species in the units overlying beds is followed by a gradual increase in marine and brackish taxa, suggesting a gradual RSL rise, and hence interseismic land-level fall

(Fig. 4.6, 4.9, 4.11). Although not as pronounced, the influx of marine and brackish diatoms above beds 5 and 7 at Tirúa is followed by an increase in freshwater taxa, suggesting RSL fall, and hence interseismic land-level rise.

Diatoms spanning two of the beds at Tirúa (beds 2 and 9) and one bed at Quidico (2) display no significant change in salinity preference or species composition, suggesting

133 CHAPTER 4 coseismic land-level change was not large enough (probably less than a few tens of centimeters) to cause an ecological change, or, in the case of bed 9, the bed may not have been not deposited by a tsunami (Fig. 4.8, 4.10, 4.12).

4.5.3 Earthquakes and land-level change

Ely et al. (2014) correlate beds 1, 2, 4, and 5 at Tirúa with recent historical tsunamigenic earthquakes in AD 2010, 1960, and most likely historical earthquakes in 1751, and 1575, respectively. Hong (2014) correlate beds 1, 2, 3, and 4 at Quidico with recent historical tsunamigenic earthquakes in AD 2010, 1960, and most likely historical earthquakes in

1835, and 1751, respectively (Fig. 4.8, 4.10, 4.12). Our diatom results show coseismic uplift or subsidence that matches historical accounts and stratigraphic evidence of coseismic deformation associated with the AD 1960, 1835, 1751, and 1575 earthquakes

(Cisternas et al., 2005; Udías et al., 2012). We also describe four previously undocumented tsunamigenic earthquakes (beds 6, 7, 8, 9) at Tirúa, correlate one of the prehistoric events with a sand bed discovered by Hong (2014) in the Quidico sedimentary sequence (bed 7), and attempt to correlate the other new events with other records of prehistoric earthquakes in the region.

4.5.3.1 Historical earthquakes

Although surveys conducted before and soon after (Ely et al., 2014), and weeks following

(Farías et al., 2010; Melnick et al., 2012) the AD 2010 earthquake measured between 0.5 and 1.0 m of coseismic uplift, we were unable to analyze diatom samples of sediment overlying the 2010 bed at Tirúa or Quidico (bed 1) because neither the river nor the tides have widely deposited sediment on bed 1 since the tsunami. However, the 2010 rupture provides critical information regarding the type of slip distributions that cause uplift at 134 CHAPTER 4 our sites. Inferred slip distributions for the 2010 earthquake show deep, landward slip on the Maule segment results in uplift at Tirúa and Quidico (Moreno et al., 2012).

Similar diatom assemblages underlying and overlying bed 2 suggest no coseismic vertical deformation at Tirúa following the AD 1960 Valdivia earthquake sequence, which included a foreshock (Mw 7.5) and a mainshock (Mw 9.5; Cifuentes, 1989; Fig. 4.1b). Our results are consistent with historical accounts and surveys that document little to no coseismic vertical deformation at Tirúa following the earthquakes (Plafker and Savage,

1970; Cifuentes, 1989) and diatom analysis by Garrett et al. (2013) that found no land- level change coincident with sand deposition. The 1960 inferred slip distribution is characterized by a shallow slip zone concentrated offshore, which resulted in widespread subsidence of the coastline south of Tirúa (Fig. 4.13; Moreno et al., 2009). The AD 1575 earthquake, thought to be the predecessor of the 1960 earthquake, likely had a similar slip distribution based on historical accounts of coastal subsidence, yet produced subsidence at Tirúa (Fig. 4.13). The discrepancy between the deformation at Tirúa in

1960 and 1575 may be the result of the Mw 7.5 foreshock that struck offshore of

Concepción. At the Quidico headlands (Fig. 4.2c), Plafker and Savage (1970) estimated up to a meter of coseismic uplift during the 1960 foreshock based on pre- and post- earthquake leveling of tides and lower growth limits of vegetation and accounts from local residents. However, we observed no change in paleoecology in sediments underlying and overlying bed 2 at Quidico, 1 km upriver from the headlands. Our results suggest that the uplift associated with the 1960 foreshock and the subsidence accompanying the mainshock created a complicated deformation pattern that resulted in no net RSL change at Tirúa and Quidico. Thus, we can infer that the shallow, offshore

135 CHAPTER 4 slip associated with Valdivia segment ruptures can result in subsidence or no vertical deformation at our sites (Plafker et al., 1970; Moreno et al., 2009).

The sudden increase in freshwater diatoms (i.e. RSL fall) above bed 3 at Quidico is consistent with historical accounts of coseismic uplift in and around Concepción during the AD 1835 (M 8.0-8.5) Maule segment earthquake (Fig. 4.12; Darwin, 1839; Lomnitz,

2004). Tirúa sections T16 and T2 contain no diatom evidence of uplift or tsunami inundation in AD 1835 (Fig. 4.8, 4.10), suggesting Quidico may have been on the southern edge of the deformation zone (Fig. 4.1b). Additionally, the absence of bed 3 in the Tirúa sedimentary sequence suggests the northward facing embayment at Quidico may result in the selective preservation of tsunamis generated by Maule segment ruptures (Fig. 4.2b; Hong, 2014).

The pronounced increase in freshwater diatoms (i.e. RSL fall) above bed 4 at Tirúa and

Quidico is consistent with historical accounts of coseismic uplift in the Concepción area associated with the AD 1751 (M ~8.5) earthquake on the Maule segment (Lomnitz,

2004). The 1751 earthquake and tsunami were highly destructive—leading to the inland relocation of Concepción—and have been compared to the AD 2010 (Mw 8.8) rupture of the Maule segment and accompanying tsunami (Lomnitz, 2004). The distinct lithologic and diatom evidence for uplift and tsunami inundation in 1751 at Tirúa and Quidico suggests that this earthquake was one of the largest to affect this part of Chile, and supports its comparison with the Mw 8.8 2010 rupture. However, it is unclear if the uplift accompanying the 2010 earthquake at Tirúa and Quidico will be similarly preserved in the sedimentary record due to the lack of post-2010 sedimentation. Similar sedimentary hiatuses following the 2010 earthquake and tsunami were described by

136 CHAPTER 4 Garrett et al. (2013) at sites on the Arauco peninsula, highlighting the importance of considering gaps in the sedimentary record when applying diatoms to reconstruct coseismic uplift and subsidence (Fig. 4.1; Shennan et al., 2009). The lack of post-2010 sedimentation may also reflect changing conditions at the Tirúa and Quidico rivers since

1751, likely due to anthropogenic modifications, such as the construction of sea walls and bridges, and the paving of the river channels.

The abrupt increase in brackish and marine diatoms (i.e. RSL rise) above bed 5 at Tirúa is consistent with historical accounts of coseismic subsidence from Puerto Saavedra to

Chiloe Island during the AD 1575 (M ~9.5) Valdivia segment earthquake, thought to be the predecessor of the giant AD 1960 earthquake (Moernaut et al., 2014). Coseismic subsidence stratigraphy and tsunami deposits recording the AD 1575 events are preserved at the Maullín River estuary (41.5° S; Cisternas et al., 2005) and very strong ground shaking during this earthquake is recorded by lacustrine turbidites (e.g., Lago

Villarica) northeast of Valdivia (39.5°S; Moernaut et al., 2014; Fig. 4.13). Quidico section

Q9 contains no diatom evidence of subsidence or tsunami inundation in AD 1575, suggesting that Tirúa may have been on the northern edge of the deformation zone in AD

1575 (Fig. 4.1b). Additionally, the absence of a sand bed associated with the AD 1575 tsunami at Quidico may reflect selective preservation of evidence for Maule-segment- generated tsunamis there.

We did not find evidence of the historical AD 1837 or 1657 earthquakes in the Tirúa and

Quidico sequences. The lack of stratigraphic or microfossil evidence for the AD 1837 earthquake and tsunami is consistent with historical accounts suggesting that the rupture occurred on the southern portion of the Valdivia segment and its effects were not

137 CHAPTER 4 as severe in localities north of Valdivia (Moernaut et al., 2014). The absence of the AD

1657 earthquake and tsunami evidence at Tirúa can be explained by the relatively small magnitude of the earthquake, described by Udías et al. (2012) as comparable to the AD

1835 Maule segment earthquake, for which evidence of its tsunami is only preserved at

Quidico (Fig. 4.12). Although the evidence is not definitive, Hong (2014) describes a discontinuous bed between sands 4 and 7 that may have been deposited by the AD 1657 tsunami, consistent with the selective preservation of relatively lower magnitude Maule earthquakes there.

4.5.3.2 Prehistoric earthquakes

Our interpretation of changes in diatom assemblages across sharp contacts inferred to mark coseismic RSL changes match the directions and relative amounts of historical coseismic land-level changes suggesting that diatom-based stratigraphic analysis can be reliably used to reconstruct earthquake history and provide some insight into megathrust slip distribution during prehistoric events.

An increase in freshwater diatoms (i.e. RSL fall) above bed 6 at Tirúa suggests coseismic uplift coincident with sand deposition. The modeled age for bed 6 deposition is AD 1457-

1575 (Table 4.4), which is constrained by the bed 5 age estimate (AD 1575) and the next oldest bed 7, which has a modeled age of AD 1443-1547 (Fig. 4.13). Coseismic uplift associated with bed 6 is indicative of a deep, landward slip pattern, typical of ruptures on the Maule segment (e.g., earthquakes in 2010, 1835, 1751). It is possible that the bed was deposited by the first earthquake in the historical record—the AD 1570 Maule segment rupture—but this is unlikely due to the 30 cm of sediment separating the bed from the

AD 1575 bed (5; Fig. 4.8). Further, the AD 1570 earthquake is not widely documented in

138 CHAPTER 4 historical records compared to the widespread accounts of the AD 1575 event, suggesting it was a much smaller rupture (Cisternas et al., 2005; Fig. 4.1b). Other regional paleoseismic studies do not record a prehistoric rupture on the Maule segment during the AD 1457-1575 time range for deposition of bed 6 (Cisternas et al., 2005; Moernaut et al., 2014). This may be evidence of a previously unidentified earthquake and accompanying tsunami, whose amount of deformation is consistent with a Maule segment rupture. Hong (2014) describes a discontinuous bed deposited in this time range at Quidico, which may correlate with bed 6 at Tirúa, but the bed was not present in section Q9, complicating correlation between sites.

We correlate bed 7 between Tirúa and Quidico based on similar radiocarbon ages, similar pattern of deformation, and bed thickness (Fig. 4.8, 4.12). The modeled age for the deposition of bed 7 at Tirúa is AD 1443-1547, while the Quidico age model shows deposition between AD 1425-1465 (Hong, 2014). A significant increase in brackish and marine planktonic diatoms (i.e. RSL rise) above bed 7 indicates coseismic subsidence coincident with sand deposition. Coseismic subsidence associated with bed 7 is indicative of a shallow, offshore slip pattern, typical of ruptures on the Valdivia segment

(e.g., 1960, 1575). Bed 7 is the thickest deposit we documented in sections T16 (15 cm thick) and Q9 (11 cm thick), suggesting the tsunami generated by this rupture was at least as large as tsunamis in 1960 and 1751, whose deposits were preserved at both sites.

Hong (2014) correlated bed 7 with an AD 1466 Valdivia segment earthquake preserved in turbidite records (Moernaut et al., 2014). Moernaut et al. (2014) concluded that the AD

1466 event was similar in size and location to the small 1737 rupture offshore of Valdivia based on the limited preservation of the ~1466 turbidite compared to the major, widespread turbidites associated with the 1960 and 1575 ruptures (Fig. 4.1b, 4.13).

139 CHAPTER 4 Although poorly documented, the 1737 earthquake is not associated with a tsunami or coseismic land-level change. Thus, the distinctive tsunami deposits and evidence for coseismic subsidence at Tirúa and Quidico for the AD 1466 event are inconsistent with a small AD 1737-style rupture; our data suggests a larger Valdivia segment rupture and extensive tsunami.

The modeled age range of bed 8 (AD 256-461) at Tirúa overlaps with age estimates for stratigraphic evidence of coseismic subsidence in AD 0-250 at Maullín (Cisternas et al.,

2005), AD 250-650 at Valdivia (Nelson et al., 2009), and an extensive lacustrine turbidite in AD 270-290 described by Moernaut et al. (2007) in Lago Puyehue, 100 km southeast of Valdivia (Fig. 4.13). However, diatom evidence of coseismic uplift associated with bed 8 is inconsistent with a Valdivia segment rupture like that implied by the widespread earthquake and tsunami evidence in the Valdivia region. Instead, the coseismic uplift and deposition of bed 8 is likely the result of a previously undocumented

Maule segment rupture.

Sand 9 (modeled age: AD 176-336) is a better candidate for correlation with the large

Valdivia rupture recorded in this time period at Maullín, Valdivia, and Lake Puyehue. We discovered little to no vertical deformation coincident with the deposition of bed 9. The giant 1960 rupture also produced little to no land-level change at Tirúa, suggesting a similar slip distribution for the AD 174-335 earthquake. In the absence of evidence of land-level change coincident with sand deposition, another possibility is a non-tsunami origin of the sand. However, the diatom composition, grain-size, and LOI values of bed 9 are similar to values for beds in sections T16, T2, and Q9 that are attributed to tsunamis, supporting a tsunami origin.

140 CHAPTER 4 4.5.4 Length of record and recurrence intervals of earthquakes

The length of the earthquake and tsunami record at Tirúa (~2000 years) is comparable to coastal stratigraphic records near Maullín (Atwater et al., 1992; Cisternas et al., 2005) and Valdivia (Nelson et al., 2009), where coastal sedimentary sequences are underlain by distinct unconformities a few hundred to a few thousand years old. Above the unconformity, Atwater et al. (1992) and Nelson et al. (2009) describe reworked tidal and floodplain mud, peat, and sand with only fragmentary evidence of coseismic land-level change or tsunami inundation. Like our record at Tirúa and Quidico, Cisternas et al.

(2005) describe a more complete sequence of earthquakes and/or tsunamis, with evidence for eight events in the last 2000 years. However, unconformities exist within this record, with over 500 years separating older events. Similarly, a distinct unconformity of ~1000 years occurs within the 8 cm of sediment separating beds 8 and 7 at Tirúa.

Atwater et al. (1992) and Nelson et al. (2009) attribute the fragmentary nature of south- central Chile coastal paleoseismic records to net emergence (RSL fall) during the late

Holocene. Field evidence in south-central Chile and ICE-5G VM5a model predictions for the region suggest RSL fell from a high of 3-8 m in the mid-Holocene to its present level

(e.g., Mitrovica and Milne, 2002; Peltier, 2004; Milne et al., 2005; Isla et al., 2012). In the last 1500 years, field observations indicate less that 1 m of RSL fall (Atwater et al.,

1992). Our diatom data from section T16 is consistent with minimal net RSL fall over the last ~1500 years; the base of section T16, which contains sediment dated to AD 100-500, contains a higher percentage of tidal flat and marine planktonic and littoral diatoms, consistent with higher RSL. By AD ~1400, the abundance of marine diatoms has

141 CHAPTER 4 decreased, and remains stable until present day, showing little to no net RSL change at

Tirúa or Quidico in the last ~600 years (Garrett et al., 2014).

Based on the five historical and two prehistoric vertical deformation and tsunami events preserved at Tirúa and Quidico in the last 600 years, the recurrence interval of large (M

>8), near-source tsunamigenic earthquakes ranges from 50 to 180 years, with an average recurrence interval of ~100 years (Fig. 4.13). A century also separates the two earthquakes and tsunamis (beds 8 and 9) preserved below the inferred unconformity at

Tirúa. A 100-year recurrence interval is consistent with the historical recurrence of destructive (M >8) earthquakes and tsunamis on the Maule segment (Lomnitz, 2004), however our records do not contain two of the ruptures (AD 1570 and 1657) used to calculate the historical interval. Instead the 100-year interval between earthquakes at our sites probably reflects their location in the area where ruptures generated on the

Maule and Valdivia segments overlap, allowing evidence of deformation during northern and southern earthquakes to be preserved. Based on our results, the recurrence interval for Maule segment ruptures large enough to be preserved in the stratigraphic record is

~170 years. The recurrence interval between Valdivia segment ruptures at our sites is

~250 years, consistent with a recurrence rate of ~280 years determined by Cisternas et al. (2005) tidal lowland sections at Maullín and by Moernaut et al. (2014) from lacustrine turbidite stratigraphy east of Valdivia.

4.6 CONCLUSIONS

1) We documented a total of nine tsunamigenic earthquakes in the sedimentary sequences at Tirúa and Quidico, including four prehistoric ruptures.

142 CHAPTER 4 2) Diatom analysis confirms previous inferences of tsunami deposition and historical accounts of land-level change at Tirúa and Quidico during historical earthquakes on the

Maule and Valdivia segments. A sudden, widespread, and lasting increase in freshwater diatoms above beds 3 and 4 is consistent with historical accounts of coseismic uplift and tsunami inundation during Maule segment earthquakes in AD 1835 and 1751, respectively. Little to no change in diatom composition above bed 2, and a sudden, widespread, and lasting increase in marine and brackish diatoms above bed 5, are consistent with historical accounts of Valdivia-segment earthquakes that report little to no vertical deformation in AD 1960 and widespread subsidence in AD 1575.

3) The same diatom methods applied to stratigraphic sections yield evidence for four prehistoric earthquakes; we identified evidence of coseismic uplift coincident with tsunami deposition in AD 1457-1575 (bed 6) and 256-461 (bed 8), and coseismic subsidence or no change coincident with tsunami deposition in AD 1443-1547 (bed 7) and 176-336 (bed 9), respectively.

4) Based on historical patterns of earthquake deformation and our diatom-based reconstructions of prehistoric coseismic land-level change, we attribute ruptures in AD

1457-1575 and AD 256-461 to the Maule segment, and ruptures in AD 1443-1547 and AD

176-336 to the Valdivia segment.

5) Our prehistory may include two previously undocumented Maule segments earthquakes (beds 6 and 8), and two earthquakes that correlate with previously described Valdivia segment earthquakes (beds 7 and 9).

6) The 100-year recurrence interval of large tsunamigenic earthquakes at our sites reflects their location in the region of overlap between the Maule and Valdivia segments, 143 CHAPTER 4 where evidence of earthquakes with both northern and southern sources is preserved.

The recurrence interval of Maule segment earthquakes (~170 years) represents only events large enough to leave a coastal stratigraphic record, and excludes smaller earthquakes as in 1837, 1657, and 1570. The recurrence interval of Valdivia segment earthquakes preserved at Tirúa and Quidico is ~250 years, consistent with recurrence interval estimates based on stratigraphy in the Maullín and Valdivia regions.

The presence of both historical and prehistoric evidence of coseismic vertical deformation in south-central Chile provided an opportunity to test the efficacy of stratigraphic and diatom reconstructions of earthquakes and tsunamis before applying them to the prehistoric record. Our results demonstrate the utility of diatom-based reconstructions of coseismic deformation in addressing questions of long-term subduction-zone behavior and segmentation along the south-central Chilean subduction zone.

4.7 ACKNOWLEDGMENTS

This work was supported by the U.S. National Science Foundation grants EAR-1145170 and EAR-1144537, National Geographic Society Research Grant 8577-08, and Chile

Fondo Nacional de Desarrollo Científico y Tecnológico (FONDECYT N° 1110848). We thank Brian Atwater, Ed Garrett, Marcelo Lagos, Andrew McConkey, Caitlin Orem, and

Daniel Pineda for collaboration in field or laboratory research, and Sonja Hausmann and

Don Charles at Academy of Natural Sciences in Philadelphia, USA for their assistance with diatom analyses. Nelson and Wesson are supported by the Earthquake Hazards

Program of the U.S. Geological Survey. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

144 CHAPTER 4

Figure 4.1: Index maps. a) Plate-tectonic setting of Chile in western South America. b) Location of the study area in central Chile, seismic segments, main tectonic features, and estimated rupture lengths of the largest historical earthquakes in south-central Chile since 1570 compiled from Lomnitz, (1970), Kelleher (1972), Comte et al. (1986), Beck et al. (1998), Melnick et al. (2009), Udías et al. (2012), and Melnick et al. (2012).

145 CHAPTER 4

Figure 4.2: a) The Tirúa and Quidico rivers are located about 160 km and 170 km south of Concepción, respectively. b) Satellite Google Earth Pro, 2014 DigitalGlobe imagery of the Tirúa River lowland. Riverbank transect A-A’ is bracketed in blue. Red dots show location of sections T16 and T2. c) Satellite Google Earth Pro, 2014 DigitalGlobe imagery of the Quidico River lowland. Core and section locations are shown with section Q9 highlighted. Described cores and sections were projected onto elevation transect B-B’.

146 CHAPTER 4

Figure 4.3: Correlated stratigraphic units in transect A-A’ at Tirúa (a) and B-B’ at Quidico (b) (see Figure 4.2b and 4.2c for transect locations). Elevations are relative to mean tidal level (MTL). Sections chosen for detailed analysis (T16, T2 and Q9) are highlighted in red.

147 CHAPTER 4

148 CHAPTER 4 Figure 4.4 (previous page): a,b) High-resolution grain-size distribution color surface plots (Φ) showing the differential volume of sediments (the percentage of total volume that each size class occupies), field photograph of section, detailed lithology, D10 values (diameter at which 10% of a sample's mass is comprised of smaller grains), and LOI (weight %) for stratigraphic columns of sections T16 and T2. Lithologic descriptions of units are based on the mean grain size values. Elevations are relative to mean tide level (MTL). Radiocarbon sampling locations and calibrated radiocarbon ages are plotted on columns. Radiocarbon age determinations from this study are in green and ages from Ely et al. (2014) are in blue. Radiocarbon sample numbers correspond to sample numbers in Table 4.3. 137Cs analysis is from Ely et al. (2014).

Figure 4.5: High-resolution grain-size distribution color surface plots (Φ) showing the differential volume of sediments (the percentage of total volume that each size class occupies), field photograph of section, detailed lithology, D10 values (diameter at which 10% of a sample's mass is comprised of smaller grains), and LOI (weight %) for stratigraphic columns of section Q9. Lithologic descriptions of units are based on the mean grain size values. Elevations are relative to mean tide level (MTL). Radiocarbon sampling locations and calibrated radiocarbon ages are plotted on columns. Radiocarbon age determinations from Hong (2014) are in orange. Radiocarbon sample numbers correspond to sample numbers in Table 4.3. 137Cs analysis is from Hong (2014).

149 CHAPTER 4

Figure 4.6: Summary of microfossil analyses for section T16. Triangle symbols on stratigraphic column show locations of numbered samples (sample numbers shown to the right of the salinity summary). Relative abundance of diatoms (showing only those species > 5% abundance) is expressed as a percent of the total count. The results of the stratigraphically constrained incremental sum-of-squares (CONISS) cluster analysis are shown beside our interpretation of paleoecological zones. Be: benthic; Pl: planktonic.

150 CHAPTER 4

151 CHAPTER 4

Figure 4.7 (previous page): Results of detrended correspondence analysis (DCA) for diatom samples from sections T16 (a), T2 (b), and Q9 (c). Samples with similar species compositions are grouped together in the DCA bi-plot and samples with statistically different species compositions plot apart (Horton et al., 2007). Paleoecological zones identified by the cluster analysis are labeled on the DCA bi-plot. Arrows show how assemblage bi-plot scores change from below to above labeled bed contacts.

Figure 4.8: Summary of evidence for land-level change and tsunami inundation for section T16. a) The ratio of freshwater to marine diatoms, the % marine and brackish diatoms, and the % marine/brackish planktonic diatoms highlight sudden changes in the dominant salinity and environmental preference of diatoms under-and overlying sand beds. The ratio of freshwater to marine diatoms is left out to highlight the change in assemblages across the beds. The % marine and brackish diatoms and % marine/brackish planktonic diatoms also highlight changes in diatom composition within tsunami sand beds. Evidence for tsunami inundation includes the high relative abundance of tidal flat diatoms within sand beds and the presence of marine planktonic diatoms within sand beds. Inferred coseismic land-level change and timing of sand bed deposition are shown.

152 CHAPTER 4

Figure 4.9: Summary of microfossil analyses for section T2. Triangle symbols on stratigraphic column show locations of numbered samples (sample numbers shown to the right of the salinity summary). Relative abundance of diatoms (showing only those species > 5% abundance) is expressed as a percent of the total count. The results of the stratigraphically constrained incremental sum-of-squares (CONISS) cluster analysis are shown beside our interpretation of paleoecological zones. Be: benthic; Pl: planktonic.

153 CHAPTER 4

Figure 4.10: Summary of evidence for land-level change and tsunami inundation for section T2. a) The ratio of freshwater to marine diatoms, the % marine and brackish diatoms, and the % marine/brackish planktonic diatoms highlight sudden changes in the dominant salinity and environmental preference of diatoms under-and overlying sand beds. The ratio of freshwater to marine diatoms is left out to highlight the change in assemblages across the beds. The % marine and brackish diatoms and % marine/brackish planktonic diatoms also highlight changes in diatom composition within tsunami sand beds. Evidence for tsunami inundation includes the high relative abundance of tidal flat diatoms within sand beds and the presence of marine planktonic diatoms within sand beds. Inferred coseismic land-level change and timing of sand bed deposition are shown.

154 CHAPTER 4

Figure 4.11: Summary of microfossil analyses for section Q9. Triangle symbols on stratigraphic column show locations of numbered samples (sample numbers shown to the right of the salinity summary). Relative abundance of diatoms (showing only those species > 5% abundance) is expressed as a percent of the total count. The results of the stratigraphically constrained incremental sum-of-squares (CONISS) cluster analysis are shown beside our interpretation of paleoecological zones. Be: benthic; Pl: planktonic.

155 CHAPTER 4

Figure 4.12: Summary of evidence for land-level change and tsunami inundation for section T2. a) The ratio of freshwater to marine diatoms, the % marine and brackish diatoms, and the % marine/brackish planktonic diatoms highlight sudden changes in the dominant salinity and environmental preference of diatoms under-and overlying sand beds. The ratio of freshwater to marine diatoms is left out to highlight the change in assemblages across the beds. The % marine and brackish diatoms and % marine/brackish planktonic diatoms also highlight changes in diatom composition within tsunami sand beds. Evidence for tsunami inundation includes the high relative abundance of tidal flat diatoms within sand beds and the presence of marine planktonic diatoms within sand beds. Inferred coseismic land-level change and timing of sand bed deposition are shown.

156 CHAPTER 4

Figure 4.13: Summary of historical and paleoseismic evidence for Valdivia and Maule segment earthquakes over the last ~2000 years. Historical records summarised by Heck (1947), Berninghausen (1962); Lomnitz (1970), Reed et al. (1988), Cisternas et al. (2005), and Udías et al. (2012); stratigraphic and biostratigraphic evidence from Cisternas et al. (2005, 2007), Nelson et al. (2009), Garrett et al. (2013, 2014); Ely et al. (2014), Moernaut et al. (2007), Moernaut et al. (2014). VS: Valdivia segment; MS: Maule segment; FS: Foreshock.

157 CHAPTER 4 TABLE 4.1. ECOLOGY OF DIAGNOSTIC DIATOM SPECIES IN SECTIONS T16, T2, and Q9

1 From Denys (1991); Marine = M; Brackish = B; Freshwater = FW. 2,3 Compiled from Rivera (1983), Rebolledo et al., (2005, 2011), Krammer and Lange-Bertalot (1986, 1988, 1991a,b), Denys (1991), Vos and de Wolf (1988, 1993), Hemphill-Haley (1995), Hartely et al. (1996), Lange-Bertalot, (2000), Witter et al. (2009). Planktonic = diatoms that float freely in the water column and do not live attached to any substrate; Tychoplanktonic = diatoms that live in the benthos, but are commonly found in the plankton; Epipelic = diatoms that live on or just below the surface of wet muddy sediments; Epiphytic = diatoms that are attached to larger plants or other surfaces; Aerophilic = diatoms that are able to survive subaerial, temporarily dry conditions.

158 CHAPTER 4 TABLE 4.2. MEAN GRAIN SIZE, D10, AND LOI OF SAND BEDS AND UNITS IN T16, T2, and Q9

159 CHAPTER 4

TABLE 4.3. RADIOCARBON AGES, TIRÚA AND QUIDICO, CHILE

1 Sample ID’s and corresponding sampling locations can be found on Figure 4.4a,b and Figure 4.5. 2 Sections with corresponding radiocarbon age locations can be found on Fig. 4.3a,b. 3 Interpretation of the stratigraphic context of the dated sample relative to the time that host unit was deposited. Maximum ages are on samples containing carbon judged to be older than a sand bed. Minimum ages are on samples judged younger than a sand bed. *Radiocarbon age from Ely et al. (2014) **Radiocarbon age from Hong et al. (2014)

160 CHAPTER 4

TABLE 4.4. MODELED AGES FOR SAND BEDS 6, 7, 8, AND 9 AT TIRÚA AND EARTHQUAKE SLIP DISTRIBUTION

1 We used maximum and minimum limiting radiocarbon ages to calculate age probability distributions for inferred prehistoric tsunami beds from Tirúa section T16 with the V-sequence feature of OxCal ( Bronk Ramsey 2008; e.g., DuRoss et al., 2011; Berryman et al., 2012) 2 Land-level change inferred from diatom based RSL reconstructions. 3 Slip distribution and rupture location inferred from prehistoric land-level change and historical earthquake patterns

161 CHAPTER 4

TABLE 4.5. DCA AXES SCORES FOR SECTIONS T16, T2, AND Q9

1 We used the multivariate statistical program MVSP to run detrended correspondence analysis. 2 Sample numbers correspond to sample numbers in Figures 4.6, 4.7, 4.9, and 4.11.

162

CHAPTER 5. CONCLUSIONS

5.1 INTRODUCTION

The overarching aim of my research was to assess the applicability of diatoms to reconstruct earthquake and tsunami records over multiple earthquake cycles (thousands of years) in central and south-central Chile. If successful, I would be able to address fundamental questions about the slip distributions, segmentation, and recurrence intervals of the largest subduction zone earthquakes and tsunamis. This chapter concludes the thesis by assessing the extent to which the initial research aims and objectives outlined in Chapter 1 have been met. I also consider future research avenues.

5.2 SUMMARY OF THESIS

1. Review the application of diatoms to paleoseismic studies in order to identify the

advantages and disadvantages of the method, and provide a resource for future

investigations.

Chapter 2 filled a knowledge gap in paleoecology by providing a detailed review of the application of diatoms to paleoseismic studies around the world. I discussed how the utility of diatoms as sea-level indicators is underpinned by the concept that characteristic assemblages have strong associations with salinity, tidal elevation, and substrate. A database of 77 relevant studies and detailed examples illustrated the application of diatoms to reconstruct RSL change associated with coseismic subsidence and uplift, and tsunami inundation. I highlighted the variability of diatom evidence for earthquake related RSL change and tsunami inundation and discussed the importance of understanding the geological and geomorphological context of sampling locations, and

163

CHAPTER 5 the processes operating within them when making paleoecological interpretations. I concluded that many of the disadvantages to diatom-based earthquake and tsunami reconstructions could be addressed by continuing to explore the modern associations of diatoms to environmental variables, thereby improving the accuracy of paleoecological interpretations. This chapter will be submitted to Earth Science Reviews in December

2014.

2. Identify suitable geomorphic environments in central and south-central Chile

that preserve geologic evidence of past earthquakes and tsunamis for detailed

stratigraphic and biostratigraphic analyses.

I selected three low-energy coastal sedimentary environments from the coast of Chile for stratigraphic investigations. In central Chile (Chapter 3), site selection was a major challenge because low-energy marsh environments are very rare on the semi-arid, rocky coastline. I selected a unique coastal lowland (Quintero) that historical maps indicated was once covered by a lagoon, and that is presently covered by some of the only marsh vegetation in central Chile. Fifty-seven cores, hand-dug pits, and exposures revealed laterally continuous sand beds in the site stratigraphy, indicating it contained a record of repeated tsunami inundation. In the temperate climate of south-central Chile (Chapter

4), low-energy marsh environments are more common. There, I targeted two estuarine marshes (Tirúa and Quidico), where a total of 69 bank exposures, pits, and cores revealed stratigraphic evidence of coseismic land-level change and tsunami inundation.

3. Characterize site sediments based on their lithology, grain size distribution, and

organic matter content.

164 CHAPTER 5 Chapters 3 and 4 described the systematic assessment of sedimentary sequences representative of site stratigraphy through lithologic, grain size, and organic content

(loss on ignition; LOI) analyses. Detailed (1-5 cm resolution) sampling of sedimentary sections clearly defined 15 sand beds at Quintero, Tirúa and Quidico; in total I performed

550 grain-size and LOI measurements. Grain-size analysis and LOI revealed a fine lag

(Low D10 values) and low organic content in sand beds, consistent with local sand sources at my sites (e.g., beach, dune, and estuarine sediments) and supporting an oceanward, and hence, tsunami source for the sands. D10 and LOI values were also consistent with values observed in known tsunami deposits in 1960 and 2010, further supporting a tsunami source for the beds. Upward fining of grain size, indicative of sediment settling out of suspension as a tsunami flow decelerates after reaching its inland extent, was found in five of the six sand beds described in Chapter 3.

Analyses of the sediments underlying and overlying the sand beds revealed a higher percentage of fine material (higher D10 values) and higher organic content, indicating the sediments formed in situ and were covered in vegetation, clearly distinguishing them from the sand beds.

4. Develop a taxonomic database of fossil diatoms in central and south-central Chile

and determine the salinity and substrate preference of fossil diatoms from each

site using existing modern diatom studies.

I identified 150 species in 165 fossil diatom samples counted in Chapters 3 and 4. A minimum of 300 diatoms were counted per sample, with each species expressed as a percentage of total diatom valves counted. I counted at least 50,000 diatoms from my

Chile investigations. Diatoms were classified by salinity (marine, brackish, and

165 CHAPTER 5 freshwater) and life-form (planktonic, epipelic, epiphytic, aerophilic) based on the ecological preferences outlined in Chilean (Rivera, 1983; Rebolledo et al., 2005, 2011) and global (Krammer and Lange-Bertalot, 1986, 1988, 1991a,b; Vos and de Wolf, 1988,

1993; Denys, 1991; Hartely et al., 1996; Lange-Bertalot, 2000) catalogs. Eighty-two diatom species that exceeded 5% of total valves counted are included in a fossil diatom database table, and were used for paleoecological interpretations (Appendix 1).

5. Explore intra- and inter-site variability in diatom evidence for RSL change

related to the earthquake deformation cycle and tsunami inundation.

I inferred a tsunami origin for a sequence of six sand beds in central Chile that exhibited characteristics described in modern tsunami deposits such as significant lateral extent

(100 to 200 m), uniform thickness, sharp (1-3 mm) lower and upper contacts, and upward fining sequences (Chapter 3). Diatom assemblages within the sand beds included anomalous marine planktonic species not present in surrounding sediments, supporting a oceanward, and hence, tsunami origin for the sands. Paleoecological changes within the sands were laterally continuous throughout the lowland, helping rule out localized processes such as tidal channel migration, liquefaction, or sand venting during earthquakes as mechanisms responsible for sand bed deposition. No changes in lithology were evident between sediment underlying and overlying the sand beds, however, diatom analyses revealed coseismic uplift coincident with sand deposition.

Diatom evidence of uplift was laterally continuous throughout the lowland, strengthening my argument for a tsunami origin for the sand beds; although storm- related barrier breaches can produce widespread sand beds with some characteristics similar to tsunami deposits, breach deposits are not coincident with coseismic uplift.

166 CHAPTER 5 Diatom analyses described in Chapter 4 confirmed previous inferences of tsunami inundation and historical accounts of coseismic land-level change during five historical earthquakes in south-central Chile in AD 2010, 1960, 1835, 1751, and 1575 and identified four prehistoric tsunamigenic earthquakes in AD ~1500, ~1450, ~350, ~250. Diatoms showed laterally continuous intra-and inter-site paleoecological changes in sediment overlying sand beds indicating coseismic uplift in AD 1835, 1751, ~1500, and ~350, subsidence in AD 1575 and ~1450, and no change in AD 1960 and ~250. Four of nine sand beds contained anomalous marine planktonic taxa. The remaining sand beds contained mixed diatom assemblages. The mixed diatom assemblage found in the sand beds is typical of tsunami deposits, because tsunamis erode, transport, and deposit marine, brackish, and freshwater sediments (with associated taxa) as they inundate coastal and inland areas. In the absence of anomalous marine planktonic taxa, we relied on the diatom evidence of widespread coseismic land-level change coincident with sand deposition to support a tsunami origin of the sand beds.

6. Compare diatom-based reconstructions with historic documentation of coseismic

deformation and accompanying tsunamis.

The presence of both observational and historical evidence of coseismic vertical deformation in south-central Chile (Chapter 4) provided an opportunity to test the efficacy of stratigraphic and diatom methods before extending them to the prehistoric record. Observational and historical accounts of land-level change and inundation by tsunamis confirmed my diatom-based reconstructions. A sudden, widespread, and lasting increase in freshwater diatoms above sand beds associated with Maule segment ruptures in AD 1835 and 1751 is consistent with historical accounts of coseismic uplift

167 CHAPTER 5 during those earthquakes. Little to no change in diatom composition above a sand bed associated with the AD 1960 Valdivia segment earthquake and tsunami is consistent with little to no observed vertical deformation during that earthquake, while a sudden, widespread, and lasting increase in marine and brackish diatoms above a bed associated with a Valdivia segment rupture in AD 1575 is consistent with historical accounts of widespread subsidence. Historically, Maule segment ruptures have produced uplift at my sites, while Valdivia segment ruptures have produced subsidence.

Witnesses describe tsunami surges flooding the study sites in AD 1960 and 2010, supporting a tsunami origin for the two youngest sand beds in my sections and providing analogues for older events. The 1960 and 2010 tsunami deposits are composed of fine sand (mean grain size = 2.20 Φ), display a low fine fraction (mean D10 = 3.38 Φ) and organic content (mean LOI = 3%), and contain anomalous marine planktonic diatoms.

The seven older sand beds in my sections display very similar sedimentologies (mean grain size = 2.26 Φ; mean D10 = 3.45 Φ; mean LOI = 3%), and two contain anomalous marine diatoms, supporting a tsunami source for the sand beds.

7. Extend the earthquake and tsunami record of central and south-central chile and

develop recurrence intervals of great subduction zone earthquakes.

In south-central Chile (Chapter 4), I applied the diatom methods for the observational and historical record to stratigraphic sections. I provided evidence for four prehistoric earthquakes; in total I identified nine tsunamigenic earthquakes.

I identified evidence of coseismic uplift coincident with tsunami deposition in AD ~1500 and ~350, and coseismic subsidence or no change coincident with tsunami deposition in

AD ~1450 and ~250, respectively. Based on historical patterns of earthquake 168 CHAPTER 5 deformation and my diatom-based reconstructions of prehistoric coseismic land-level change, I attribute ruptures in AD ~1500 and ~350 to the Maule segment, and ruptures in AD ~1450 and ~250 to the Valdivia segment. Chapter 4 illustrated the efficacy of diatom-based reconstructions of coseismic deformation in addressing questions of long- term subduction-zone behavior and segmentation along the south-central Chilean subduction zone.

I attributed the relatively short (100-year) recurrence interval of large tsunamigenic earthquakes in south-central Chile to the location of my studies in the region of overlap between the Maule and Valdivia segments, where evidence of earthquakes with both northern and southern sources can be preserved. The recurrence interval of the four

Maule segment earthquakes we document (~170 years) represents only events large enough to leave a coastal stratigraphic record, and excludes smaller earthquakes in 1837,

1657, and 1570. The recurrence interval of Valdivia segment earthquakes preserved at my sites is ~250 years, consistent with recurrence interval estimates based on stratigraphy in the Maullín and Valdivia regions

In central Chile (Chapter 3), I described evidence for six previously undocumented tsunamigenic earthquakes between ~6200 and 3600 cal yr BP. My interpretations were supported by diatom analyses showing anomalous marine planktonic diatoms within the sand beds and uplift associated with five of the six beds. The interval between great earthquakes ranged from ~200 to ~650 years with an average recurrence interval of

~500 years, comparable to recurrence intervals at other subduction zones thought to preserve evidence of only unusually large (> Mw 8.5) earthquakes and tsunamis.

However, the recurrence interval for the three most recent historical earthquakes (AD

169 CHAPTER 5 1822, 1906, and 1985) and low tsunamis in central Chile is much shorter (~80 years) suggesting that the events recorded in the Quintero lowland were unlike those of the historical record. I concluded that the best historical analog for the prehistoric tsunamigenic earthquakes preserved at Quintero is the unusually large AD 1730 earthquake and high tsunami, and that basing hazard assessments on only the most recent, well-documented, but smaller earthquakes and tsunamis in 1822, 1906, and 1985 overlooks the higher hazard posed by less frequent, larger events.

8. Explore the Holocene RSL history of sites and the limits it imposes on the

preservation of earthquake and tsunami evidence.

The ~6200-3600 cal yr BP record of earthquakes and tsunamis in central Chile (Chapter

3) coincides with a period of gradual RSL rise between ~8000 and ~4000 yrs BP that culminated in a highstand of ~1.4 m at ~4200 cal yr BP. During this period, the eustatic contribution to sea level was dominant and net sea-level rise provided the accommodation space necessary to preserve earthquake and tsunami evidence.

Following the highstand, accommodation space was limited, resulting in brief or fragmentary paleoseismic records.

The relatively complete record of late Holocene earthquakes and tsunamis in south- central Chile (Chapter 4) over the last ~600 years is likely the result of relatively stable

RSL over that time, while the fragmentary prehistoric record is typical of falling RSL and reworking of sediments. My results indicating relatively stable RSL over the last 600 years contradict glacio-isostatic adjustment models that predict a highstand at ~8000 cal yr BP and falling RSL since. This may be the result a complex interplay between

170 CHAPTER 5 eustatic, tectonic and local factors, in addition to poorly understood glacio-isostatic uplift of the coastline following the Last Glacial Maximum.

I have published Chapter 3 (Dura et al., 2014) in Quaternary Science Reviews. Chapter 4 will be submitted to Geological Science of America Bulletin in Spring 2015.

5.3 FUTURE CONSIDERATIONS

My thesis represents an attempt to identify coseismic deformation and tsunami inundation in coastal sedimentary sequences using diatom-based RSL reconstructions.

In this sense the initial thesis aims stated in Section 1.2 have been met. However, there remain many opportunities expand the temporal and spatial resolution of paleoseismic studies in Chile, and elsewhere.

In Spring 2015, I will travel to Chile to continue the search for suitable geomorphic environments for the preservation of earthquake and tsunami evidence. Thus far, evidence for prehistoric tsunamigenic earthquakes in central Chile has only been documented at my site, Quintero. It is critical expand the spatial resolution of paleoseismic evidence in central Chile in order to better understand the magnitude of past events. Pollen studies have identified additional marshes along the central Chile coast that may contain Holocene sedimentary sequences similar to those described at

Quintero. An extensive anomalous shell bed and sand layer indicative of high tsunamis near Quintero is another point of interest. I have received a NOSAMS award for 35 radiocarbon dates to attempt to constrain the age of the shell bed.

In south-central Chile, there is a need to explore more sites north of Concepción. The spatial resolution for Valdivia segment ruptures south of my sites is relatively good,

171 CHAPTER 5 however no paleoseismic studies have been conducted along the coast between

Concepción and Valparaíso. Again, the spatial resolution of paleoseismic evidence and deformation estimates must be improved in order to better understand the nature of prehistoric ruptures.

My major research goal for the future is to quantify coseismic deformation through the application of diatom-based transfer functions. Presently, the modern diatom dataset in

Chile does not adequately reflect diatom variability in a variety of geomorphic environments. Thus, issues arise due to a lack of good modern analogues for fossil assemblages, restricting my thesis to qualitative interpretations. I can potentially address this issue by expanding the modern diatom training set, and continuing to explore the modern associations of diatoms to environmental variables, thereby improving paleoecological interpretations.

In addition, I will explore the response times of diatoms to changes in RSL. In diatom- based reconstructions of coseismic coastal deformation, we assume that diatoms are responding instantaneously RSL change. Thus, diatoms immediately overlying sharp lithologic contacts are interpreted to represent the post-earthquake elevation of the marsh. However, species response times may not be instantaneous, and diatoms incorporated into sediments above sharp contacts may lag in their response to the environmental change, complicating deformation reconstructions because both coseismic and postseismic deformation will be incorporated into estimates (Hawkes et al., 2010). One means to test the response times of diatoms to RSL change is to simulate rapid sea-level rise. I have the opportunity to conduct diatom analysis on surface samples from a marsh in Oregon that has recently been reclaimed and flooded by tides

172 CHAPTER 5 following a long period of being dammed. My diatom analysis will provide insight into the response of microfossils to sudden changes in salinity and tidal inundation.

I also hope to explore the potential for comparison of diatom-based coseismic deformation estimates with predictions from elastic dislocation models. Improved spatial resolution, precision, and confidence in diatom-based reconstructions of coseismic deformation may enable dislocation modeling for Chilean subduction zone earthquakes similar to that conducted at Cascadia (Hawkes et al., 2011; Wang et al.,

2013), and will provide further insight into the spatial variability of strain accumulation and release along the megathrust.

I am also interested in further investigating the composition of diatoms within tsunami deposits. My results show that diatom evidence for tsunami deposition is highly variable between sites, and highlight the importance of understanding the geological and geomorphological context of a site, and the processes operating within the site, when inferring a tsunami source for a sand bed. I would like to explore modern tsunami deposits in a wide range of geomorphic environments to assess the influence of site conditions on the composition of diatom assemblages within the deposits. By expanding the database of modern analogues, we can improve our interpretations of prehistoric events.

Another possible research avenue is the evaluation of the taphonomic character (e.g. fragmentation) and grading of diatom valves within a tsunami deposit. These proxies may provide insight into the flow conditions of tsunami surges. Although diatom fragmentation and grading has been observed in modern tsunami deposits and been used to support a tsunami origin for prehistoric anomalous sand beds, the proxies are

173 CHAPTER 5 not well defined and to my knowledge, have not been tested in a laboratory setting. I hope to culture diatoms in the laboratory and conduct flume experiments to determine how the taphonomy of diatoms is affected by a variety of flow conditions and substrates, and if grading is evident in diatom valves as they settle out of suspension.

Grading of sand and silt particles within tsunami deposits can also shed light on the flow conditions of a tsunami surge (Jaffe et al., 2011). In a numerical techniques and applications course, I explored a model used for calculating tsunami flow speed from the thickness and grain size distribution of tsunami deposits. This model has been tested on modern tsunami deposits, but its application to prehistoric deposits remains dubious, due to the unknown site conditions and sediment sources thousands of years into the past. I am interested in further exploring the application of this model to prehistoric deposits, in the hopes that flow speeds can be calculated at multiple sites in order to gain a better perspective on the relative size of past tsunamis.

The completeness of the tsunami record also needs to be investigated. How often are tsunami deposits completely erased from the sedimentary record? I had the opportunity to survey the 2010 tsunami at Tirúa, which covered the majority of the lowland in a 5cm-

15 cm sheet of sand, and in subsequent trips to Chile have observed almost complete erosion of the deposit in some portions of the lowland. This field season, I plan to conduct another systematic survey of the deposit in order to determine the extent of the erosion. This investigation of a known tsunami deposit associated with coseismic uplift of the coast will provide valuable insight into possible missing events in our prehistoric tsunami records.

174 CHAPTER 5 Distinguishing between sand beds deposited during the largest storms and by tsunamis is also a topic I would like to explore. Like tsunamis, the identification of storm-surge deposits is based on anomalous sand deposits in an environment where they are unusual, such as lakes and marshes, and the diatom species present in the storm layer reveal some information about the source of the sediment (Switzer and Jones, 2008;

Parsons et al., 2008; Tuttle et al., 2004). However, in many cases it is impossible to differentiate a storm deposit from a tsunami deposit based on diatoms alone. The strongest evidence for sediment deposition by tsunami that diatoms can provide is evidence for land-level change coincident with sand bed deposition. Expanding the database of modern analogues will help identify any characteristics that distinguishing characteristics of diatom assemblages in storm and tsunami overwash deposits, but it is likely that a multi-proxy approach combining of microfossil, geochemical, anthropological, and historical-analog data will be required.

Finally, the long-term relative sea-level histories of coastlines must be considered when interpreting diatom evidence of past earthquakes and tsunamis. Along coastlines that record a mid-Holocene highstand (Mitrovia and Milne, 2002; Peltier, 2004; Milne et al.,

2005) stratigraphic evidence of earthquakes and tsunamis is often found during a limited period when the eustatic contribution to sea level was dominant and net sea-level rise created low-energy depositional environments that provided the accommodation space necessary for preservation (e.g., Dura et al., 2011; Brill et al., 2013). Thus, fossil diatom assemblages will reflect distinctly different paleoenvironments than are found in the present day. In addition, higher past relative sea levels must be considered when inferring the magnitude of tsunamis based on the presence of sand beds, as low tsunamis

175 CHAPTER 5 that may not have a significant impact on the coast today may have been able to inundate lowlands in the past.

I hope to apply the diatom and grain-size methods I have explored in my dissertation to other subduction zones sites around the world. Paleoseismic studies are lacking along the Alaska-Aleutian megathrust and the southern Japan trench, providing an opportunity for novel diatom-based reconstructions of coseismic deformation and tsunami inundation there.

176

APPENDIX

A1. ECOLOGY OF DIAGNOSTIC DIATOM SPECIES (EXCEEDEDING 5% OF TOTAL VALVES COUNTED), CENTRAL AND SOUTH-CENTRAL CHILE

177

1 From Denys (1991); Marine = M; Brackish = B; Freshwater = FW. 2,3 Compiled from Rivera (1983), Krammer and Lange-Bertalot (1986, 1988, 1991a,b), Vos and de Wolf (1988, 1993), Hemphill-Haley (1995), Hartely et al. (1996), Atwater and Hemphill-Haley (1997), Sherrod (1999), Lange-Bertalot, (2000), Rebolledo et al., (2005, 2011),; Witter et al. (2009). Planktonic = diatoms that float freely in the water column and do not live attached to any substrate; Tychoplanktonic = diatoms that live in the benthos, but are commonly found in the plankton; Epipelic = diatoms that live on or just below the surface of wet muddy sediments; Epiphytic = diatoms that are attached to larger plants or other surfaces; Aerophilic = diatoms that are able to survive subaerial, temporarily dry conditions.

178

A2. FULL DIATOM COUNTS FROM QUINTERO, TIRÚA, AND QUIDICO, CHILE 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.4 4.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a ct n u p ri t

a l cu vi a N 3.9 0.7 1.0 6.3 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a 17.0 18.3 l a h p ce o ch yn rh

a l cu vi a N 0.0 0.0 0.0 1.5 5.6 0.0 0.0 0.0 0.0 0.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a l l e n e t o t cryp

a l cu vi a N 1.2 1.3 1.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 s i l va o b su s s i e n o l p i D s 2.2 2.3 2.9 1.1 1.3 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 n a g e l e

a l cu i t n e D 2 . s 6.9 6.2 8.0 0.9 2.1 1.1 0.7 0.0 2.5 2.8 2.0 1.3 h 11 t 16.7 13.2 13.4 12.3 17.6 24.0 23.7 i l o yt h p

d n a s s e yt h p ryso h C r e mb u n

e 5 6 7 8 9 0 2 3 4 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 Sa P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 179 s 2.5 2.7 3.9 8.9 0.0 3.8 0.0 18.5 28.0 24.4 21.3 61.0 55.8 29.9 85.0 47.4 16.0 25.8 15.2 21.1 rmi o f e a e f f co ra o h Amp a 0.0 0.0 0.0 0.0 1.3 2.3 0.0 0.0 0.5 3.8 3.7 9.7 2.3 0.5 0.0 0.0 0.0 0.0 0.0 0.0 t a n n i p a l l re si ro u a St a 8.4 9.0 4.2 4.0 4.0 6.3 2.3 2.6 0.0 0.0 0.3 9.7 4.6 1.4 1.0 1.4 0.0 0.0 l 10.5 12.3 ru e b b i g a i d o l a p o h R 1.2 1.3 0.3 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a t a st co vi re b a ri a l u n n Pi s s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 5.8 3.5 3.0 2.4 0.0 0.0 0.0 0.0 0.0 0.0 ri a e n i l a i zsch t i N 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 4.9 2.5 1.7 3.3 0.0 0.0 0.0 0.0 0.0 0.0 a t a l o ce n a l a i zsch t i N r e mb u n e 5 6 7 8 9 0 2 3 4 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 Sa P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4

180 s 2.5 2.7 3.9 8.9 0.0 3.8 0.0 18.5 28.0 24.4 21.3 61.0 55.8 29.9 85.0 47.4 16.0 25.8 15.2 21.1 rmi o f e a e f f co ra o h Amp a 0.0 0.0 0.0 0.0 1.3 2.3 0.0 0.0 0.5 3.8 3.7 9.7 2.3 0.5 0.0 0.0 0.0 0.0 0.0 0.0 t a n n i p a l l re si ro u a St a 8.4 9.0 4.2 4.0 4.0 6.3 2.3 2.6 0.0 0.0 0.3 9.7 4.6 1.4 1.0 1.4 0.0 0.0 l 10.5 12.3 ru e b b i g a i d o l a p o h R 1.2 1.3 0.3 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a t a st co vi re b a ri a l u n n Pi s s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 5.8 3.5 3.0 2.4 0.0 0.0 0.0 0.0 0.0 0.0 ri a e n i l a i zsch t i N 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 4.9 2.5 1.7 3.3 0.0 0.0 0.0 0.0 0.0 0.0 a t a l o ce n a l a i zsch t i N r e mb u n e 5 6 7 8 9 0 2 3 4 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 Sa P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4 P4

181 a l 6.9 0.0 0.0 3.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 l e n e t o t cryp a l cu vi a N m 1.7 0.0 5.3 0.0 4.5 3.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.7 u l 10.9 rvu a p ma e n o h mp o G a t 0.0 8.3 3.7 3.0 0.0 0.0 1.3 3.7 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a n d a a mi e h t i Ep s i l 6.3 0.0 2.0 6.3 8.0 8.8 0.0 0.0 0.0 0.0 3.2 2.6 0.0 0.0 0.0 3.0 2.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 va o b su s s i e n o l p i D s n 2.6 0.0 5.7 0.0 2.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 2.0 0.0 0.0 a 15.7 18.9 16.8 14.8 22.0 g e l e a l cu i t n e D s h 7.7 9.1 9.6 0.3 0.0 5.2 9.9 6.4 2.0 5.0 3.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 t i 16.5 25.5 14.3 13.7 l o yt h p d n a s s e yt h p ryso h C r e mb u n

b e 1 2 3 4 5 6 7 8 9 0 2 3 4 5 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 Sa P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 182

0.0 0.0 3.0 1.0 4.5 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 0.0 Rhopalodia gibba Rhopalodia s i 0.0 4.3 3.7 6.0 8.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 d 12.5 ri vi a ri a l u n n Pi 0 a . t 5.3 0.0 5.7 9.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a 11 12.9 st co vi re b a ri a l u n n Pi m 1.7 0.0 1.3 5.7 0.0 0.0 2.3 3.3 0.0 5.3 0.0 1.0 0.0 9.0 8.4 3.0 3.9 3.5 2.3 2.6 2.0 0.0 2.6 0.0 2.3 u l u st ru f a i zsch t i N 0 . a i 4.0 5.0 8.5 2.3 2.7 0.0 0.0 0.0 0.0 0.0 2.3 4.2 0.0 0.0 3.0 0.0 1.7 0.0 3.7 3.3 b 11 i 20.8 13.7 10.2 17.0 h mp a a i zsch t i N s 6.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.9 0.0 0.0 4.0 1.7 1.3 3.0 0.0 0.0 si n e n o b i l a l cu vi a N i t 0.0 0.0 5.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.6 0.0 2.0 0.0 0.0 0.0 rd a h n i re a l cu vi a N r e mb u n

b e 1 2 3 4 5 6 7 8 9 0 2 3 4 5 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 Sa P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 183

9 . i i 6.9 0.0 3.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 h 11 10.7 10.4 23.4 65.0 55.0 31.0 10.2 66.0 60.0 t smi s s i e n o l p i D a 1.7 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 3.0 2.6 0.0 0.0 3.6 2.3 5.3 2.3 0.0 4.0 l l si u p a l l e ymb C a n 0.0 0.0 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.3 1.0 0.0 0.0 0.0 0.0 2.3 2.7 2.7 0.0 2.7 2.3 9.0 0.0 a i 39.9 n i h g e n me a l l e t o ycl C s 3.6 0.0 0.0 0.0 0.0 0.0 0.0 2.3 0.0 0.0 0.0 0.3 0.0 0.0 4.2 8.6 6.8 3.0 3.0 4.0 5.0 5.3 12.9 25.0 30.2 rmi o f e a e f f co ra o h Amp 0.0 1.3 6.0 3.5 2.4 2.6 0.0 0.0 0.0 0.0 0.3 0.0 3.0 2.6 4.6 0.0 3.0 3.0 0.0 2.3 2.3 0.0 2.7 0.0

26.4 a t a n n i p a l l re si ro u a St 3.3 0.0 3.7 5.0 5.5 4.4 2.6 2.7 6.2 4.2 0.0 0.0 3.5 3.0 2.6 2.7 2.7 2.0 2.3 2.3 2.3 2.7 5.0 21.3 16.7 a l ru e b b i g a i d o l a p o h R r e mb u n

b e 1 2 3 4 5 6 7 8 9 0 2 3 4 5 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 Sa P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1

184

0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 0.0 0.0 0.0 0.0 0.0 0.0 2.9 5.0 4.2 8.0 8.0 0.0 2.0 4.0 0.0 0.0 s 14.3 e p vi re b s s e h t n a n Ach a i 0.0 0.0 0.0 1.7 0.0 0.0 0.0 0.0 3.0 4.0 2.2 0.0 0.0 0.4 0.0 5.6 4.6 9.7 7.3 5.3 5.9 4.3 d 10.7 10.7 10.0 rme e t n i

. r va s s e p vi re b s s e h t n a Ach m 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.3 0.0 0.0 0.0 1.7 0.0 0.0 0.0 0.0 0.0 u 10.2 l u t ca i l e d m m u i d i h t o n a Pl a 6.3 0.0 2.0 2.7 0.0 0.0 5.3 7.3 0.0 0.0 0.0 1.0 0.0 0.0 0.0 0.0 0.0 8.7 2.0 2.0 2.3 0.0 0.0 0.0 l 10.7 a h p ce o ch yn rh a l cu vi a N 0 . 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.0 3.0 2.5 0.3 9.3 9.6 9.0 9.0 4.3 6.3 3.6 2.7 3.0 m m 11 10.0 10.7 19.2 ru e f i d o n ma g si yro G r e mb u n

b e 1 2 3 4 5 6 7 8 9 0 2 3 4 5 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 Sa P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1

185 i i g 0.0 0.0 0.0 0.0 0.0 0.0 4.3 5.3 0.8 0.0 0.0 0.0 0.0 0.0 0.0 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 o l ssf i e w ra si o ssi a l a h T s 0.0 0.0 0.0 0.0 0.0 0.0 7.6 7.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.3 0.7 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 ri st cu a l ra si o ssi a l a h T a 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 1.3 0.0 0.0 0.0 4.8 3.4 0.0 0.0 0.0 0.0 0.0 1.7 0.0 0.0 t ca l su a i l ra Pa 7 . a t 0.0 0.0 6.7 6.5 7.2 8.0 0.4 3.6 8.6 0.0 a 11 l 10.3 43.8 49.0 14.1 35.7 14.8 24.8 28.0 18.7 21.7 17.2 29.7 18.2 10.3 cu sci a f a ri a l u b a T 6 a . l u 0.0 0.0 0.0 0.0 0.0 0.0 2.0 0.0 0.0 0.0 0.7 2.0 0.4 0.4 4.8 7.0 7.7 5.3 0.0 0.0 2.3 t 11 12.1 20.0 20.0 a ri st a l l re ri Su m 1.7 0.0 0.0 0.0 0.0 0.0 0.0 2.7 0.0 0.0 0.0 0.0 0.4 0.0 4.2 3.0 6.2 0.0 0.0 4.0 4.3 4.3 u l 10.3 15.7 25.1 l e t scu s s i e n cco o C r e mb u n

b e 1 2 3 4 5 6 7 8 9 0 2 3 4 5 5 6 7 8 9 0 1 2 3 4 l 1 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 ...... mp 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 Sa P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1 P1

186

0.0

0.0

0.0

2.3

0.0

0.0

0.0

0.0

0.0

0.0

0.0

5.1

1.0

3.3

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.9

0.0

0.0

0.0

0.0

1.7

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

3.6

0.0

0.0

m

u

l

u

v

r

pa

a

m

e

n

o

ph

m

o

G

3.3

1.8

1.6

4.2

0.0

3.4

5.0

3.2

3.3

2.5

2.5

6.8

5.3

3.1

0.0

0.0

0.0

0.0

0.0

0.0

3.1

2.6

4.3

5.7

1.3

2.7

3.7

0.4

1.9

5.4

3.3

8.6

9.1

3.7

3.0

0.5

0.6

0.3

0.4

0.8

0.7

1.0

1.0

0.7

1.0

0.0

11.5

10.1

11.5

10.1

11.2

s

i

l

a

v

do

e

psu

s s

i

e

n

o

pl

i

D

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.9

0.0

1.3

4.3

0.0

1.2

0.0

0.0

0.0

0.0

0.0

0.0

1.3

0.0

0.0

0.0

0.0

0.0

0.0

s

i

l

bti

su

a

l

cu

ti

n

e

D

0.0

0.0

1.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.3

0.0

5.8

0.0

0.0

0.0

0.0

0.0

2.5

3.7

7.4

1.6

6.6

9.2

8.2

3.0

5.3

0.9

1.3

3.0

3.0

5.0

6.0

2.6

5.1

0.0

9.0

0.0

1.3

a

l

13.7

16.1

23.7

21.9

12.0

14.3

12.0

11.0

10.0

l

si

pu

s s

i

e

n

o

i

sm

o

C

a

7.1

1.8

3.8

4.9

8.3

9.7

6.1

0.7

0.0

0.0

0.0

0.0

6.5

5.6

3.4

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

l

tu

n

ce

a

pl

s s

i

e

n

cco

o

C

0.0

0.0

0.0

1.3

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

5.8

0.0

1.1

2.8

0.0

0.0

1.1

0.0

0.0

0.0

0.0

0.0

2.3

4.0

0.0

0.0

0.3

0.3

0.0

0.0

4.3

1.6

0.0

0.0

0.0

2.7

0.0

0.0

0.0

1.0

3.6

2.3

1.0

0.0

0.0

12.0

sa

do

pi

a

l

a

l

u

n

i

v

a

C

9.5

1.5

8.0

5.5

1.7

6.6

7.6

6.6

8.5

3.5

0.0

5.3

3.8

6.6

9.6

0.9

2.3

0.0

0.8

0.0

1.2

2.2

0.0

0.0

0.0

0.3

0.0

0.0

3.0

0.0

0.0

0.0

0.0

0.0

0.0

1.0

0.3

1.0

0.5

0.0

0.0

0.8

1.0

1.7

0.7

3.6

1.3

3.7

13.7

10.0

12.6

s

i

l

a

v

o

a

r

o

ph

m

A

r

be

m

u

n

e

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

5

1

0

9

8

7

6

2

1

0

9

7

6

4

3

2

1

0

9

8

4

3

2

1

pl

.6

.6

.6

.6

.6

.5

.5

.5

.5

.5

.5

.5

.5

.5

.5

.4

.4

.4

.4

.4

.4

.4

.4

.4

.4

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

m

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

a

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1 S

187

2.7 1.0 1.6 6.9 9.6 2.7 0.0 5.5 0.0 2.3 7.3 3.5 4.0 5.6 4.9 7.3 5.3 4.6 6.9 6.3 6.0 8.5 7.3 3.7 3.4 0.6 1.2 0.0 0.0 0.0 0.0 0.0 5.3 8.2 2.8 5.1 0.0 0.0 1.5 2.7 0.0 0.0 0.0 5.5 1.0 2.3 0.0 10.3 12.0 11.8 12.0 m u l stu u r f a i sch tz Ni 2.0 1.0 0.0 0.0 1.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a bi i ph m a a i sch tz Ni 0.0 0.0 7.6 3.0 2.9 0.0 0.0 5.3 4.7 3.0 0.0 0.0 0.0 1.0 1.0 5.1 2.8 6.8 6.0 0.0 0.0 3.6 1.6 4.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.0 1.3 0.0 0.0 5.8 3.8 0.0 1.6 1.4 0.0 3.9 5.1 3.2 1.2 m 13.4 11.7 u n pi l a m u di i Ne 0.0 0.0 4.6 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a l a ph ce o ch n y h r a l cu i v Na 0.0 0.0 0.0 7.6 6.9 6.3 0.4 0.4 0.7 0.0 1.0 0.0 1.0 0.4 0.0 3.0 4.6 4.0 4.0 7.2 5.0 8.7 0.0 2.0 2.9 4.1 6.5 4.0 1.7 0.9 0.0 0.9 0.0 0.0 0.0 0.0 5.6 3.8 4.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.6 1.6 1.5 2.1 i 10.1 r ca a l cu i v Na 0.0 0.0 7.5 3.0 3.0 3.0 2.1 2.5 1.0 5.0 3.5 2.5 3.0 4.2 5.1 2.2 0.0 0.0 0.0 1.3 0.0 1.9 0.0 2.8 4.0 2.3 1.7 1.2 1.0 2.5 3.4 2.2 0.0 0.0 0.5 0.5 2.7 0.0 0.0 0.0 2.2 0.0 3.7 1.0 0.0 0.0 0.0 0.0 13.7 12.3 12.3 ca ti u m a l co ti u L 0.0 1.0 0.0 0.3 0.0 0.0 0.0 0.0 1.0 1.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 7.0 7.1 6.8 1.7 0.0 1.0 5.7 5.0 0.0 2.5 7.7 8.1 4.1 4.3 1.3 2.8 1.1 0.0 6.5 9.4 2.2 6.8 7.8 7.2 5.6 7.4 8.3 5.6 3.6

11.0 15.2 27.0 21.7 10.0 a l l e g n o bl o a i v e y a r a K 0.0 0.7 8.1 8.0 9.0 0.0 0.0 2.0 9.4 7.6 3.0 3.0 1.3 0.0 8.0 2.3 0.0 0.0 0.0 1.6 0.0 2.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.5 2.8 1.8 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0 s s 13.0 10.7 12.3 13.6 10.3 10.3 10.1 y x o i ph m a a i sch tz n a H r be m u n e 1 2 3 4 8 9 0 1 2 3 4 6 7 9 0 1 2 6 7 8 9 0 1 5 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 pl .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .6 .6 .6 .6 .6 m 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 a S T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1

188

0.0 1.6 0.0 0.0 4.3 0.3 0.0 1.3 0.7 2.0 0.7 0.0 0.0 2.4 4.7 0.0 0.0 6.7 7.2 6.0 3.3 0.0 0.7 3.5 4.1 3.8 8.4 6.0 6.4 2.0 3.8 2.2 1.1 3.3 1.9 0.0 0.0 1.3 0.0 0.0 0.0 0.0 3.2 2.4 0.0 0.0 0.0 1.6 0.0 2.7 s 12.7 i m r o f e a e f f co a r o ph m A 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 2.2 0.9 1.7 3.2 3.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.8 0.0 i i n a m r o n s s u cl cy o n cti A 3.0 0.7 0.0 0.0 0.0 0.3 1.3 0.0 0.0 6.0 2.0 1.0 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 sa do o n a i r a l u n n Pi 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.8 0.0 0.0 0.0 1.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 i dti ste r e g a l a i r a l u n n Pi 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.3 0.3 6.7 4.7 3.7 8.8 1.0 2.8 0.0 0.0 4.7 3.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.0 0.0 0.0 0.0 a 15.2 13.0 12.3 15.3 21.2 10.8 10.1 di e m r te n i a i r a l u n n Pi 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.7 1.2 2.9 0.0 0.0 0.0 0.0 0.0 0.0 2.7 1.3 2.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 s i l bi o n g i a i r a l u n n Pi 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.6 0.0 0.0 0.0 0.0 8.7 8.0 1.3 1.9 0.9 1.1 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 10.7 14.1 10.1 ta sta co i v e br a i r a l u n n Pi r be m u n e 1 2 3 4 8 9 0 1 2 3 4 6 7 9 0 1 2 6 7 8 9 0 1 5 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 pl .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .6 .6 .6 .6 .6 m 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 a S T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1

189

1.5

3.2

5.4

6.8

9.6

1.7

2.9

1.7

5.8

4.0

5.3

6.3

0.0

0.0

0.0

5.7

0.0

0.0

0.0

0.0

2.2

1.6

1.9

0.0

2.3

3.0

1.0

1.8

2.5

1.9

0.0

3.6

9.1

0.0

0.0

0.0

0.0

6.6

7.6

0.0

0.0

0.0

0.0

0.0

7.5

8.7

11.3

15.1

11.8

10.8

10.1

m

u

l

tu

ca

i

l

de

m

u

di

i

th

o

n

a

Pl

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.9

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

7.7

5.0

0.0

0.3

1.6

2.0

ta

ta

u

m

m

co

a

i

sch

tz

Ni

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

3.0

2.3

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.6

4.7

pta

e

l

l

y

ph

a

l

cu

i

v

Na

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

2.8

1.8

5.3

9.2

5.8

1.5

0.0

0.0

0.0

1.1

3.5

5.8

4.9

0.8

2.6

1.5

0.6

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

a

n

i

r

g

e

r

pe

a

l

cu

i

v

Na

1.8

1.2

1.0

0.0

0.0

0.0

0.0

0.7

0.0

0.0

0.0

5.3

5.3

6.6

0.0

0.0

0.0

0.0

0.0

0.0

1.2

1.4

3.1

8.4

5.4

1.6

1.3

0.0

4.0

2.3

5.5

3.4

2.9

5.2

0.0

0.0

0.0

1.3

0.0

0.7

3.0

2.0

0.0

0.4

3.7

6.3

2.0

1.3

0.0

12.9

11.8

cta

n

ci

a

l

cu

i

v

Na

2.4

0.0

3.8

1.0

0.0

2.6

3.4

5.1

4.8

1.8

5.7

0.7

4.5

1.0

0.0

0.0

4.4

9.0

8.5

5.9

6.6

2.5

4.1

1.3

2.5

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.7

0.0

1.0

0.7

i

i

16.4

16.6

th

i

sm

s s

i

e

n

o

pl

i

D

0.9

0.9

3.5

1.3

4.6

3.4

5.8

2.9

6.8

7.6

6.6

0.0

0.0

0.0

2.8

5.8

4.5

5.0

0.0

0.0

0.0

1.7

4.6

0.0

0.0

0.0

0.0

3.7

0.0

0.0

1.3

0.0

0.0

0.0

0.0

0.0

1.2

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

15.4

ta

u

n

i

m

di

o

e

n

s s

i

e

n

cco

o

C

r

be

m

u

n

e

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

7

6

5

4

3

2

1

0

9

8

5

1

0

9

8

7

6

2

1

0

9

7

6

4

3

2

1

0

9

8

4

3

2

1

pl

.6

.6

.6

.6

.6

.5

.5

.5

.5

.5

.5

.5

.5

.5

.5

.4

.4

.4

.4

.4

.4

.4

.4

.4

.4

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

m

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

6

a

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1

T1 S

190

0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.9 2.6 2.6 2.0 0.0 0.0 2.9 1.5 a l l e r i r su s s i e n i ph l e D 0.0 2.9 0.0 0.0 0.0 0.0 0.0 0.0 3.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0.0 1.5 0.8 0.0 1.0 3.0 0.0 0.0 1.3 0.0 0.0 0.0 0.6 0.0 2.3 4.4 3.4 2.2 0.0 5.0 1.0 0.3 0.7 1.6 2.9 4.3 1.2 2.4 3.4 2.3 3.9 2.6 3.2 5.1 e ppa i k s s i e n i ph l e D 0.0 2.0 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.3 0.0 0.0 0.0 0.0 1.9 0.3 0.6 0.0 0.0 1.9 0.0 0.0 0.0 1.2 0.0 0.0 0.0 2.5 0.0 0.0 0.0 0.0 0.0 0.0 1.0 3.8 0.0 0.0 s s u i r a n se s s u ch pty o n cti A 8.0 4.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.0 1.2 0.0 0.0 0.0 0.0 1.6 0.0 0.0 0.9 0.6 3.4 2.5 4.9 6.3 3.5 4.5 3.3 1.9 3.4 2.3 2.8 5.1 6.9 7.8 7.6 6.6 9.7 9.3 1.3 0.0 0.0 0.9 a 10.7 di e m r te n i

. r a v s s pe i v e br s s e th n a ch A 4.7 3.9 1.3 4.6 2.3 3.0 6.9 8.0 4.2 5.4 7.1 1.3 0.0 0.0 0.0 2.0 1.3 1.6 0.0 4.0 0.0 0.0 3.8 2.3 0.0 0.0 0.0 0.0 0.0 0.5 1.3 0.7 0.0 0.0 0.0 0.0 0.0 0.0 2.0 1.6 0.0 2.1 0.0 s 12.0 12.0 14.5 18.7 21.7 30.6 15.9 11.4 si n de i v e l a l l e n o i bl y r T 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.3 0.0 0.0 0.0 0.0 0.0 0.8 1.4 2.0 1.1 1.8 4.2 1.3 0.9 1.7 0.3 0.0 4.4 3.3 a l tu a i str a l e r i r u S 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 4.2 2.5 2.8 0.0 0.0 2.0 0.0 2.3 1.9 3.5 2.2 2.5 7.7 8.9 4.4 4.8 2.8 2.8 1.1 0.0 0.0 1.5 2.0 2.9 0.0 0.0 0.0 7.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 i 11.3 i n sso i v e br a di o l pa o h R r be m u n e 1 2 3 4 8 9 0 1 2 3 4 6 7 9 0 1 2 6 7 8 9 0 1 5 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 pl .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .6 .6 .6 .6 .6 m 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 a S T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 191 a r 8.0 7.8 1.6 0.7 1.7 1.0 5.4 2.1 4.6 0.3 0.0 0.3 0.3 0.0 0.0 2.0 1.0 0.0 0.0 3.0 3.7 1.3 0.0 0.0 0.0 0.0 0.9 0.6 0.9 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0.0 0.0 0.0 0.0 2.4 0.0 2.3 0.3 1.0 0.9 3.6 e n te a r so a si a l a h T ca 3.7 8.5 1.3 1.0 1.0 0.0 7.5 1.3 3.6 0.7 0.0 1.3 0.0 0.0 0.0 0.7 0.7 0.0 0.0 1.3 2.3 0.7 0.0 0.0 0.0 0.0 0.3 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.2 0.0 0.0 1.4 0.0 0.0 0.0 0.0 0.0 0.0 1.3 0.0 0.0 0.0 4.2 i f ci pa a r so a si a l a h T ta 5.0 8.8 0.7 1.6 1.3 2.3 3.7 7.0 5.3 0.7 1.2 2.5 8.3 6.8 8.6 9.0 9.0 5.9 3.8 2.5 8.0 5.1 4.2 9.9 9.6 8.7 3.1 5.8 4.6 2.9 4.3 9.6 9.5 7.4 5.3 a l 21.8 18.1 20.1 17.3 13.3 14.0 10.1 10.9 11.0 13.0 14.3 10.3 10.0 13.1 13.2 16.4 u g n a a r so a si a l a h T s 3.3 4.2 1.3 0.3 0.0 2.7 4.6 5.5 4.6 0.0 0.7 0.0 0.0 1.2 1.8 0.7 1.0 0.0 0.6 3.4 1.8 0.0 1.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.8 0.0 o r ce i ph m a s s i e n o ph a h R ta 3.6 2.6 4.3 4.0 4.0 4.3 3.6 4.3 8.3 9.0 8.6 6.3 5.9 7.0 8.2 8.8 6.5 9.1 3.8 3.3 3.4 8.6 6.6 9.3 9.8 7.4 8.6 11.3 11.8 15.9 11.8 10.9 13.4 14.6 10.8 13.1 21.1 10.7 12.3 14.2 17.2 16.7 17.5 15.4 17.3 11.5 11.6 14.6 13.6 17.3 14.0 ca l su a i l a r Pa 8.2 4.0 4.0 4.0 4.0 0.0 0.0 5.9 8.7 4.7 5.1 6.5 2.0 4.3 2.9 8.2 7.1 3.6 4.3 3.6 9.4 6.8 6.1 6.2 6.5 ca i 12.0 11.4 10.1 29.7 28.3 25.1 11.1 10.1 14.2 13.1 11.4 12.9 15.4 17.7 11.6 28.0 24.3 23.1 30.8 19.8 12.3 10.4 10.0 11.9 17.5 15.2 f ci pa a r o ph pe O ca 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 6.2 3.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 i r 13.1 a g n u h ta n do ppo i H a 0.7 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.9 1.3 1.8 3.4 2.0 3.3 3.8 0.0 0.0 0.0 0.0 0.0 0.0 5.3 0.0 e a m g py a ci a l l a F s 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.7 2.3 6.7 3.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.3 1.4 1.7 3.8 2.8 1.7 7.7 0.0 0.0 0.8 0.0 0.0 0.8 5.8 6.1 3.9 3.2 1.1 2.0 1.9 3.5 7.3 0.0 bu m bo s s i e n o pl i D r be m u n e 1 2 3 4 8 9 0 1 2 3 4 6 7 9 0 1 2 6 7 8 9 0 1 5 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 pl .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .4 .4 .4 .4 .4 .4 .4 .4 .4 .4 .5 .5 .5 .5 .5 .5 .5 .5 .5 .5 .6 .6 .6 .6 .6 m 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 a S T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1 T1

192 a 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 3.3 6.8 7.8 0.0 0.0 0.5 0.0 0.0 0.0 0.0 0.0 0.0 5.2 4.2 n i 16.1 l bsa su a i r a l i g a r F a 0.0 2.8 3.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.7 4.9 2.3 5.0 1.6 2.9 0.0 0.0 0.0 di e m r te n i a ti o n u E s i l 3.1 2.2 8.4 9.6 6.8 2.0 0.0 7.4 6.4 9.9 6.9 6.6 8.3 2.5 4.7 3.1 0.5 1.7 6.8 4.2 3.4 7.1 4.8 2.8 2.4 5.0 2.3 2.7 1.7 a 12.5 v do e psu s i e n o pl i D 0.9 6.6 0.0 0.0 0.0 0.0 0.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 pta i scr de a l l be m y C a l l 0.0 0.6 0.0 0.0 2.4 0.0 0.0 0.0 1.8 3.3 9.5 2.0 0.0 1.7 0.0 0.0 0.4 2.8 1.5 3.7 0.6 1.7 1.6 3.1 0.4 9.1 3.0 3.2 23.9 19.3 si pu s i e n o i sm o C m 2.5 9.7 5.7 0.9 5.3 0.6 2.6 8.5 0.8 1.4 0.0 3.6 1.5 1.9 3.1 0.8 0.0 0.0 0.0 0.5 1.2 2.5 2.5 2.0 1.4 1.6 0.0 0.5 u l 11.0 14.3 l ci ba s i e n o l a C r be m u n e pl 1 2 3 4 6 7 8 9 0 1 3 4 5 6 7 8 9 0 1 2 3 4 5 9 0 1 2 3 4 5 m .0 .0 .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .3 a S T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2

193

1.2

0.0

0.0

2.9

2.4

0.0

0.0

0.0

1.2

1.1

4.4

1.5

3.4

2.5

3.6

1.9

3.6

2.2

2.3

3.5

8.5

0.0

3.1

6.2

7.8

4.0

1.6

0.0

0.0

i

11.2

r

ca

a

l

cu

i

v

Na

0.0

1.5

0.0

4.3

2.4

0.0

0.0

0.0

0.0

2.7

1.2

0.0

0.0

0.0

0.0

9.0

8.9

4.1

6.0

0.0

1.5

2.1

8.8

7.3

1.2

0.6

1.5

1.2

0.6

6.6

ca

ti

u

m

a

l

co

ti

u

L

7.7

7.5

5.3

3.1

4.3

2.7

1.1

2.0

5.7

4.7

6.5

5.3

0.0

0.0

2.8

0.0

0.0

2.4

1.5

0.0

0.0

1.8

0.0

0.0

4.7

0.9

4.4

a

13.1

15.0

14.6

l

l

e

g

n

o

bl

o

a

i

v

e

y

a

r

a

K

4.0

0.7

2.3

2.5

4.3

7.8

7.9

1.5

2.2

5.7

5.2

3.4

0.5

0.0

0.0

0.0

0.0

4.1

4.9

3.5

9.7

8.2

2.5

5.9

7.1

6.9

5.8

8.1

1.9

4.7

s

y

x

o

i

ph

m

a

a

i

sch

tz

n

a

H

0.0

0.0

1.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.5

2.1

0.0

0.8

0.0

0.0

0.0

0.0

0.0

0.0

11.6

11.6

12.0

11.3

m

u

l

u

v

r

pa

a

m

e

n

o

ph

m

o

G

r

be

m

u

n

e

pl

5

4

3

2

1

0

9

5

4

3

2

1

0

9

8

7

6

5

4

3

1

0

9

8

7

6

4

3

2

1

m

.3

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

a

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2 S

194

1.6 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.4 1.4 0.0 0.0 0.0 0.0 0.8 0.0 0.8 0.0 0.0 2.2 4.3 5.9 8.3 1.8 1.9 0.2 3.0 ta 11.3 sta co i v e br a i r a l u n n Pi i i 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.7 2.8 8.8 0.0 0.0 0.0 0.0 0.0 0.0 l de n a br a i r a l u n n Pi 4.1 6.6 6.5 1.8 1.2 0.6 2.0 1.6 5.3 4.0 7.2 8.9 4.1 5.3 2.5 2.5 0.8 0.0 0.0 4.0 2.7 8.8 3.7 6.8 6.9 3.1 3.9 2.6 2.2 0.0 m u l stu u r f a i sch tz Ni 5.0 7.9 0.9 1.5 1.8 2.4 2.2 0.6 0.0 1.2 3.5 0.0 2.2 2.8 1.7 0.0 0.0 2.4 4.2 6.5 8.8 7.8 4.7 5.7 2.3 2.0 0.0 11.3 10.3 10.3 a l co ti n o f a i sch tz Ni 2.8 4.7 5.0 2.7 3.9 8.3 4.2 2.5 6.2 5.2 0.0 1.1 1.9 0.0 0.0 0.0 0.0 0.0 0.0 2.4 2.7 0.0 0.9 3.1 0.0 0.0 2.5 0.0 0.2 0.0 m u n pi l a m u di i Ne r be m u n e pl 1 2 3 4 6 7 8 9 0 1 3 4 5 6 7 8 9 0 1 2 3 4 5 9 0 1 2 3 4 5 m .0 .0 .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .3 a S T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2

195

0.0

0.0

0.0

1.4

1.2

0.0

0.0

0.0

0.7

4.2

6.4

4.0

7.5

0.8

1.7

4.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.9

0.0

0.0

0.0

0.0

0.9

s

n

e

u

str

n

co

a

r

si

o

r

u

ta

S

0.0

0.5

0.0

0.0

2.0

0.0

0.0

0.0

2.2

1.1

0.8

4.4

6.0

2.5

1.1

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

ta

a

i

str

i

v

e

br

a

r

si

o

r

u

sta

do

e

Psu

0.7

1.2

5.5

1.4

6.7

7.9

7.6

3.1

1.2

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.7

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.3

19.4

11.4

sa

do

o

n

a

i

r

a

l

u

n

n

Pi

0.5

1.7

3.6

7.9

5.1

8.1

9.6

7.1

6.4

4.2

4.8

0.0

0.0

0.0

0.0

1.5

0.0

0.0

2.6

5.6

0.0

0.0

1.6

2.5

0.0

0.0

0.0

0.0

0.0

0.0

a

di

e

m

r

te

n

i

a

i

r

a

l

u

n

n

Pi

r

be

m

u

n

e

pl

5

4

3

2

1

0

9

5

4

3

2

1

0

9

8

7

6

5

4

3

1

0

9

8

7

6

4

3

2

1

m

.3

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

a

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2 S

196

3.5 7.2 6.5 4.9 9.3 5.6 4.2 0.0 6.2 4.3 5.5 1.6 3.0 3.4 4.4 2.0 9.6 8.2 2.0 0.8 1.5 0.0 0.0 0.0 4.3 6.1 0.0 0.0 1.7 cta 10.6 n ci a l cu i v Na 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.9 1.9 5.4 5.5 3.2 2.0 1.1 0.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 m u r e f di o n a m g si o r y G i 0.9 1.3 0.0 6.4 0.0 0.0 1.4 1.3 0.0 0.6 1.6 0.6 1.3 0.0 0.0 8.6 9.0 6.8 5.9 3.6 2.3 1.0 0.0 0.0 0.0 1.2 0.0 7.5 6.0 7.9 i th i sm s s i e n o pl i D 1.9 1.9 2.5 6.4 7.5 5.0 8.7 7.2 2.4 2.1 0.0 1.4 4.1 5.3 9.6 2.2 4.8 2.3 2.3 0.0 4.2 0.0 0.0 0.6 0.0 3.5 4.3 4.2 8.2 s i 11.4 m r o f e a e f f co a r o ph m A 1.6 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.3 0.0 0.0 0.0 0.0 1.5 6.4 5.9 6.5 4.0 2.4 1.1 1.2 0.9 0.0 0.0 2.4 3.9 5.5 1.5 6.7 i i n a m r o n s s u cl cy o n cti A r be m u n e pl 1 2 3 4 6 7 8 9 0 1 3 4 5 6 7 8 9 0 1 2 3 4 5 9 0 1 2 3 4 5 m .0 .0 .0 .0 .0 .0 .0 .0 .1 .1 .1 .1 .1 .1 .1 .1 .1 .2 .2 .2 .2 .2 .2 .2 .3 .3 .3 .3 .3 .3 a S T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2 T2

197

s s

u

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.2

0.0

0.0

1.7

1.0

2.3

0.8

0.0

0.0

1.3

0.0

0.0

0.0

0.0

2.2

0.0

1.2

0.9

0.9

0.0

0.6

2.8

i

r

a

n

se

a

s s

di

u

e

ch

m

r

pty

te

o

n

i

n

.

r

cti

a

A

v

s s

3.5

4.2

0.0

0.0

0.0

0.0

0.0

0.6

1.0

2.3

4.8

4.2

8.8

7.5

9.6

0.0

2.8

0.0

1.1

0.0

2.6

0.0

0.0

0.0

0.0

5.2

0.0

0.0

3.8

pe

i

13.7

v

e

br

s s

e

th

n

a

ch

A

s

si

1.2

6.5

2.9

2.1

2.8

4.7

5.4

0.0

1.5

1.1

0.0

1.5

0.5

0.0

3.0

2.2

7.6

6.0

9.5

9.4

1.8

0.0

1.7

2.1

4.0

2.8

6.0

n

14.1

13.6

12.6

de

i

v

e

l

a

l

l

e

n

o

i

bl

y

r

T

i

i

n

2.1

0.0

0.0

0.0

0.0

0.7

1.5

0.0

2.7

1.8

1.7

0.8

7.1

2.0

2.8

0.0

5.6

7.9

5.6

0.0

0.0

9.5

4.5

1.5

0.0

0.0

0.0

12.8

11.9

12.3

sso

i

v

e

br

a

di

o

l

pa

o

h

R

m

0.0

0.0

0.0

2.5

0.0

0.0

0.0

0.0

0.5

2.7

0.0

0.0

0.0

0.0

0.0

3.4

5.6

2.8

0.0

0.0

0.0

0.0

6.3

0.0

0.0

0.0

5.8

6.5

1.9

2.2

u

l

tu

ca

i

l

de

m

u

di

i

th

o

n

a

Pl

r

be

m

u

n

e

pl

5

4

3

2

1

0

9

5

4

3

2

1

0

9

8

7

6

5

4

3

1

0

9

8

7

6

4

3

2

1

m

.3

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

a

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2 S

198

ta

a

3.0

7.7

6.5

4.6

9.4

1.9

3.1

9.1

6.4

4.8

6.2

5.9

5.3

7.3

1.1

1.3

5.2

7.1

9.8

9.8

8.1

9.4

6.5

6.6

l

13.2

12.2

10.8

11.5

13.5

11.3

u

g

n

a

a

r

so

a

si

a

l

a

h

T

ta

5.4

6.2

7.1

6.3

3.1

3.4

7.7

5.2

9.3

7.0

9.7

8.2

3.7

3.2

6.7

7.6

5.6

5.9

7.1

5.7

3.6

5.3

4.1

ca

11.7

11.8

10.8

11.6

10.5

11.6

10.1

l

su

a

i

l

a

r

Pa

ca

i

0.5

2.7

0.0

2.9

1.6

0.0

1.7

4.6

4.2

6.1

9.6

4.6

6.0

4.8

5.8

9.9

6.6

2.0

1.3

2.7

2.4

9.0

3.6

3.9

4.6

4.3

6.0

6.6

f

10.3

18.2

ci

pa

a

r

o

ph

pe

O

s

5.4

2.7

2.3

0.0

0.0

0.3

0.0

0.0

0.0

3.4

1.2

3.6

0.8

1.4

1.9

0.0

0.0

1.6

0.3

0.8

0.0

0.0

1.3

3.1

1.2

1.5

0.9

0.9

0.0

6.0

bu

m

bo

s

i

e

n

o

pl

i

D

s

i

l

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.6

3.4

8.9

1.3

2.6

0.0

0.0

0.0

1.6

4.2

0.0

2.4

3.3

1.2

0.3

3.8

bti

su

a

l

cu

ti

n

e

D

e

0.0

2.0

0.0

0.0

0.0

0.0

0.0

0.9

0.0

3.4

1.6

1.5

1.3

3.1

3.9

0.0

0.0

2.2

0.0

1.9

4.6

3.5

9.7

4.5

4.1

2.7

3.6

4.3

4.1

3.8

ppa

i

k

s

i

e

n

i

ph

l

e

D

r

be

m

u

n

e

pl

5

4

3

2

1

0

9

5

4

3

2

1

0

9

8

7

6

5

4

3

1

0

9

8

7

6

4

3

2

1

m

.3

.3

.3

.3

.3

.3

.2

.2

.2

.2

.2

.2

.2

.1

.1

.1

.1

.1

.1

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

a

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2

T2 S

199

3.3 4.6 6.5 3.2 0.0 2.5 3.3 4.7 7.4 5.1 3.6 1.9 2.8 2.8 1.0 2.4 3.7 5.9 5.0 8.5 5.7 1.9 6.2 4.8 7.3 3.7 3.0 0.0 2.4 2.4 4.4 5.6 6.5 3.0 2.0 2.9 3.3 14.2 13.6 s s i l a v do e psu s s i e n o pl i D 0.0 7.5 4.0 0.0 3.0 4.7 5.0 3.4 0.0 0.0 0.0 0.6 1.3 0.6 0.3 4.8 0.0 0.7 4.6 1.0 0.0 0.0 0.0 1.2 1.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 s n a g e l e a l cu ti n e D a 1.5 8.4 3.2 1.8 4.4 4.1 9.7 2.4 3.9 3.6 5.3 1.6 1.2 1.0 5.2 1.6 1.4 1.7 1.5 1.0 0.5 2.2 2.1 1.6 2.9 2.6 0.3 2.2 2.7 5.1 1.8 2.5 3.6 6.8 1.6 0.5 1.0 0.0 l l 12.6 si pu s s i e n o i sm o C

6.1 7.8 0.0 0.0 4.2 0.0 2.4 1.6 4.2 0.0 0.0 0.0 0.9 0.0 0.0 0.0 2.5 2.1 2.0 3.5 2.1 2.5 2.2 2.1 3.0 0.0 2.0 1.7 1.1 0.0 1.9 2.4 0.0 3.8 0.0 1.1 0.0 0.0 0.0 m u l l ci ba s s i e n o l a C 0.0 0.0 0.0 0.0 0.0 0.0 1.5 0.0 0.0 0.0 4.2 0.0 0.0 0.3 1.0 0.0 0.0 1.2 3.6 3.0 0.0 0.0 2.2 0.0 0.0 0.0 0.0 0.0 1.4 0.0 0.8 1.8 1.6 1.8 0.9 1.6 0.8 1.2 2.3 s i l a v o a r o ph m A r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57

200

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

1.9

0.0

0.0

0.0

0.0

0.0

0.7

0.9

4.2

1.6

1.9

2.2

8.2

4.7

3.6

1.2

2.2

3.3

2.5

3.0

0.0

0.6

2.3

0.0

ca

ti

u

m

a

l

co

ti

u

L

3.3

3.7

2.8

5.2

2.5

1.5

2.2

4.1

3.0

2.4

1.9

0.0

0.0

3.5

1.3

1.4

1.4

0.0

0.5

0.0

0.0

0.5

2.1

2.7

8.4

6.5

4.1

2.8

0.6

1.2

0.6

0.9

1.5

1.1

0.0

1.9

1.8

1.2

0.5

a

l

l

e

g

n

o

bl

o

a

i

v

e

y

a

r

a

K

0.0

0.0

0.0

0.0

1.9

1.2

1.6

3.2

1.9

2.7

1.9

1.4

0.0

2.4

3.0

1.6

1.9

4.1

3.4

3.0

1.3

2.1

4.1

4.5

2.3

1.5

2.8

2.2

0.0

1.8

3.9

3.4

4.7

2.9

3.6

3.2

2.8

3.2

0.5

s

y

x

o

i

ph

m

a

a

i

sch

tz

n

a

H

0.0

0.0

0.5

2.5

5.2

4.7

0.0

0.9

1.1

0.0

1.4

0.0

0.0

2.3

0.0

0.0

1.1

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

2.7

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

pta

u

r

e

a

pr

a

ti

o

n

u

E

a

0.0

0.0

0.0

3.0

4.3

2.4

0.0

0.0

0.8

0.0

0.0

0.0

0.0

1.7

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

0.0

di

e

m

r

te

n

i

a

ti

o

n

u

E

r

be

m

u

n

e

pl

m

a

Q9.57

Q9.56

Q9.54

Q9.52

Q9.48

Q9.47

Q9.46

Q9.45

Q9.43

Q9.41

Q9.39

Q9.37

Q9.36

Q9.35

Q9.32

Q9.31

Q9.30

Q9.29

Q9.28

Q9.27

Q9.26

Q9.24

Q9.23

Q9.22

Q9.21

Q9.20

Q9.19

Q9.18

Q9.17

Q9.16

Q9.12

Q9.11

Q9.10

Q9.09

Q9.08

Q9.07

Q9.06

Q9.05

Q9.01 S

201

7.9 3.2 0.6 3.8 4.8 4.0 3.8 0.0 3.3 6.6 5.0 0.0 0.0 0.0 0.0 1.8 0.0 0.0 2.6 0.0 6.8 5.5 4.9 0.0 0.0 0.0 0.0 6.6 2.2 5.0 3.5 2.4 1.9 0.0 1.5 13.0 24.4 21.0 23.3 ta a i str i v e br a r si o r u sta do e Psu 0.0 0.0 0.0 2.5 0.0 0.0 0.9 0.0 0.0 5.7 6.2 2.5 2.8 4.3 1.3 2.7 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.5 2.6 3.1 2.5 3.6 1.4 2.1 1.3 7.4 7.1 1.1 0.0 0.0 0.0 a di e m r te n i a i r a l u n n Pi 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.1 2.7 0.9 1.6 1.5 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.4 3.5 2.6 0.0 2.2 1.2 1.4 1.8 1.3 3.8 1.5 0.8 1.5 0.5 0.0 ta sta co i v e br a i r a l u n n Pi m 1.3 0.0 1.5 0.6 0.0 4.4 2.1 1.9 4.2 7.8 7.1 5.3 4.4 1.9 0.6 2.7 1.4 0.5 8.7 1.3 1.1 2.2 5.1 2.4 1.7 1.2 3.1 1.9 1.8 2.2 0.0 0.0 6.2 5.6 0.5 1.8 0.7 0.0 u 10.6 l stu u r f a i sch tz Ni a l 0.0 0.0 0.9 0.0 0.0 0.0 1.5 0.6 2.7 7.2 9.5 5.0 3.8 1.5 1.0 3.3 4.8 0.0 6.9 8.2 1.8 1.9 1.4 5.8 3.2 2.0 0.0 3.7 3.0 1.2 3.2 0.0 0.0 7.1 6.5 0.8 1.5 0.5 0.0 co ti n o f a i sch tz Ni r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57 202

3.1 0.0 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.4 0.0 0.0 0.0 0.0 1.3 0.0 2.4 3.7 5.1 2.0 0.0 1.4 1.9 7.1 3.8 4.7 7.3 0.0 0.0 9.0 1.8 6.8 0.0 sp. a i r a l l ci a B s i 6.6 9.8 4.5 1.2 5.5 3.3 3.3 1.8 0.9 2.5 4.1 2.5 1.0 1.5 0.9 1.6 1.7 1.2 2.1 1.6 3.5 7.2 7.8 3.2 7.7 3.5 3.2 2.8 0.6 6.5 5.2 2.3 3.4 0.0 12.4 10.9 13.1 10.6 10.8 m r o f e a e f f co a r o ph m A 0.0 0.0 0.0 0.0 0.0 2.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.8 1.6 0.0 0.0 0.0 0.0 0.0 4.4 3.0 1.5 2.0 2.0 0.6 1.5 2.2 4.4 0.9 0.6 0.9 3.0 0.5 2.4 2.0 i i n a m r o n s u cl cy o n cti A 1.4 0.0 1.9 6.6 2.5 2.4 0.0 3.9 0.0 1.8 0.0 2.2 0.0 1.0 0.0 1.8 6.0 4.0 2.0 7.9 0.0 0.0 3.8 1.2 4.6 1.7 2.1 3.8 1.5 3.5 3.3 0.0 8.0 9.5 s 10.7 11.3 10.5 11.6 10.9 n e u str n co a r si o r u ta S 0.0 1.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 5.7 6.2 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.2 0.0 0.0 0.0 1.8 0.0 0.9 0.0 0.0 0.0 0.0 0.0 0.0 0.0 a l u r bbe i g a di o l pa o h R r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57

203 a 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.9 1.2 1.0 6.2 5.5 6.8 7.8 1.9 1.3 4.7 1.7 0.0 0.0 2.1 2.4 3.2 5.7 0.9 2.8 5.7 4.5 5.1 4.3 n i r g e r pe a l cu i v Na 5.4 0.9 1.8 2.5 7.2 6.2 2.7 2.8 6.5 6.1 7.3 4.8 1.9 1.0 5.5 6.6 3.0 6.1 4.1 7.4 8.0 7.7 6.0 8.3 5.9 3.3 0.8 0.2 0.0 15.7 18.1 12.9 10.4 11.4 14.1 12.1 10.5 10.1 10.5 cta n ci a l cu i v Na 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.6 0.6 0.0 0.0 0.0 1.3 0.0 0.0 0.0 0.0 1.2 1.0 1.0 1.8 1.6 1.9 0.0 0.0 1.2 1.7 2.0 5.8 3.6 3.0 3.8 3.8 2.7 1.2 6.8 4.0 1.0 0.0 a n i l bsa su a i r a l i g a r F 7.9 7.2 2.8 1.9 6.6 3.3 5.0 3.4 4.5 2.1 0.0 3.8 6.0 3.4 7.4 4.8 3.7 3.8 4.0 2.0 6.0 4.4 4.3 0.0 0.0 3.8 0.0 3.7 4.4 3.3 1.9 1.8 1.6 0.0 1.9 6.0 8.1

10.8 13.2 a r e n te a ci a l l a F 1.0 0.9 0.0 2.5 0.0 1.5 2.7 0.0 0.0 0.6 0.6 1.9 3.5 4.6 6.8 3.6 6.0 3.3 3.6 5.2 7.1 6.6 4.6 1.4 0.0 0.0 1.5 3.1 3.3 3.8 1.1 0.0 1.3 0.0 0.0 0.0 0.0 0.0 3.0 ta pa ci r o f a ci a l l a F 1.5 4.9 5.5 2.5 0.0 0.0 2.4 3.7 1.2 0.0 0.6 2.8 3.5 4.6 1.9 3.3 4.6 6.1 4.0 1.5 2.9 7.4 3.5 4.4 3.0 2.6 9.3 4.3 2.5 0.0 1.4 2.1 2.8 1.5 4.9 6.5 5.3 4.2 4.8 i i th i sm s s i e n o pl i D r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57

204 s 1.3 4.0 5.8 0.0 0.0 7.6 7.7 7.5 4.5 3.6 3.0 4.4 8.9 9.6 2.6 3.9 2.1 1.6 1.7 1.0 2.1 1.9 0.5 1.9 6.7 3.2 1.5 2.6 0.8 1.2 3.0 1.8 0.9 9.2 7.1 2.2 1.8 1.5 0.0 si n de i v e l a l l e n o i bl y r T 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.9 3.2 2.6 2.9 2.2 3.6 2.4 1.2 3.8 1.8 0.0 0.0 0.0 0.0 0.0 ta 13.1 a n i m cu a a di o l pa o h R i i 0.8 1.4 1.8 3.2 0.0 0.0 1.2 0.0 0.9 2.4 2.1 1.6 0.0 0.0 0.6 5.2 5.1 3.8 2.3 1.5 3.4 3.0 4.9 5.4 6.1 8.9 2.4 0.9 3.5 2.8 2.4 3.3 n 10.5 14.0 11.7 13.6 11.5 14.1 13.3 sso bi e br a di o l pa o h R 5.5 3.7 8.3 9.6 8.6 5.4 6.8 5.2 5.3 4.5 4.4 3.4 0.0 3.2 2.1 6.6 0.0 2.9 2.5 1.9 0.0 0.0 3.2 0.0 3.1 0.0 3.3 2.4 0.0 0.0 0.0 0.0 1.4 2.8 4.2 5.3 16.4 13.2 11.4 m u l tu ca i l de m u di i th o n a Pl

8.4 0.9 1.5 3.8 1.8 0.0 0.0 1.2 5.4 1.8 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.7 7.9 3.1 2.2 1.4 4.9 1.6 1.2 0.0 5.1 6.4 4.7 1.9 2.1 4.4 0.0 0.0 0.0 2.3 2.7 1.5 pta e l l y ph a l cu i v Na r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57

205

0.5 6.1 9.5 7.6 7.4 8.7 7.1 5.4 2.7 5.0 4.7 5.9 3.9 2.7 6.0 4.0 2.6 1.7 4.2 5.5 3.2 6.1 5.6 3.2 2.3 4.0 4.7 3.6 2.2 9.4 6.6 3.6 5.2 5.2 5.3 6.1 7.8 10.8 11.3 ta a l u g n a a r so a si a l a h T s o 1.5 0.6 3.7 2.5 0.0 1.1 0.0 2.5 1.8 0.6 0.9 0.0 0.0 1.2 0.0 0.0 3.0 1.9 0.0 0.0 0.0 0.0 0.0 2.8 3.0 2.6 0.0 2.0 0.0 0.0 0.0 1.5 0.0 0.9 0.0 0.0 0.0 0.0 0.0 r ce i ph m a s i e n o ph a h R ta 1.8 4.3 6.8 8.0 7.4 8.4 6.3 4.2 3.3 7.9 8.9 9.6 7.4 9.4 8.1 6.3 4.6 2.2 2.1 3.0 4.6 3.5 3.5 2.3 4.7 2.0 3.0 3.6 3.5 4.4 6.6 3.3 4.9 4.6 3.0 3.9 3.3 22.3 13.9 ca l su a i l a r Pa 4.3 7.2 4.6 8.4 4.7 4.7 4.4 5.7 2.7 2.1 6.6 4.1 8.5 4.8 1.6 2.0 0.7 1.6 0.8 3.0 0.7 0.0 0.0 0.0 0.0 2.2 2.1 2.2 0.9 3.5 0.9 2.2 2.5 2.3 1.0 1.8 ca i 13.4 10.5 14.5 f ci pa a r o ph pe O s 2.3 1.4 4.9 3.2 4.8 1.8 5.6 3.4 2.4 2.1 1.2 4.7 4.1 1.9 9.3 7.0 5.8 6.1 2.0 1.7 2.9 2.5 4.9 3.7 2.2 2.0 3.2 0.0 0.0 0.0 0.0 0.0 2.5 0.9 0.0 0.0 0.0 0.0 0.0 bu a m di e bo m s r i e te n n i o

. pl r i a D v s 0.0 0.0 0.0 1.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.2 8.5 4.9 0.0 0.0 1.3 1.1 1.4 3.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 2.1 0.0 0.0 0.0 4.6 1.8 0.0 0.0 pe i v e br s e th n a ch A r be m u n e pl m a S Q9.01 Q9.05 Q9.06 Q9.07 Q9.08 Q9.09 Q9.10 Q9.11 Q9.12 Q9.16 Q9.17 Q9.18 Q9.19 Q9.20 Q9.21 Q9.22 Q9.23 Q9.24 Q9.26 Q9.27 Q9.28 Q9.29 Q9.30 Q9.31 Q9.32 Q9.35 Q9.36 Q9.37 Q9.39 Q9.41 Q9.43 Q9.45 Q9.46 Q9.47 Q9.48 Q9.52 Q9.54 Q9.56 Q9.57

206

BIBLIOGRAPHY

Aaby, B., 1986. Palaeoecological studies of mires. In: Handbook of Holocene Palaeoecology and Palaeohydrology (ed. Berglund, B.E.), The Blackburn Press, Caldwell, New Jersey, pp.145– 164.

Admiraal, W., 1984. The ecology of estuarine sediment-inhabiting diatoms. Progress in Phycological Research 3, 269-322.

Allen, J. R. L., 2000. Morphodynamics of Holocene salt marshes: a review sketch from the Atlantic and Southern North Sea coasts of Europe. Quaternary Science Reviews, 19, 1155 - 1231.

Amspoker, M.C., and McIntire, C.D., 1978. Distribution of diatoms associated with sediments in Yaquina Estuary, Oregon. Journal of Phycology, v. 14, p. 387-395.

Arcos, M. E. M., 2012. A Holocene sedimentary record of tectonically influenced reduced channel mobility, Skokomish River delta, Washington State, USA. Geomorphology, 177, 93- 107.

Atwater, B. F., 1987. Evidence for great earthquakes along the outer coast of Washington State. Science 236, 942–944. doi:10.1126/science. 236.4804.942.

Atwater, B. F., Nunez, H. J., and Vita-Finzi, C., 1992. Net late Holocene emergence despite earthquake-induced submergence, south-central Chile, in: Impacts of Tectonics on Quaternary Coastal Evolution, Y. Ota, A. R. Nelson and K. R. Berryman (Editors), Quaternary International Vol. 15/16. Pergamon, Oxford, 77–85.

Atwater, B.F., 1992b. Geologic evidence for earthquakes during the past 2000 years along the Copalis River, southern coastal Washington. Journal of Geophysical Research, v. 97, p. 1901- 1919.

Atwater, B. F., & Hemphill-Haley, E., 1997. Recurrence intervals for great earthquakes of the past 3,500 years at northeastern Willapa Bay, Washington. U.S. Geological Survey Professional Paper, (1576), 1-108.

Atwater , B.F., Furukawa, R., Hemphill-Haley, E., Ikeda, Y., Kashima, K., Kawase, K., Kelsey, H.M., Moore , A.L., Nanayama, F., Nishimura, Y., Odagiri, S., Ota, Y., Park, S., Satake, K., Sawai, Y., Shimokawa, K., 2004. Seventeenth-century uplift in eastern Hokkaido , Japan. The Holocene, 14, 489-501.

Atwater, B. F., Uri, S. Buckley, M., Halley, R.S., Jaffe, B.E., López-Venegas, A. M., Reinhardt, E.G., Tuttle, M.P., Watt, S. and Wei, Y., 2012. Geomorphic and stratigraphic evidence for an

207

unusual tsunami or storm a few centuries ago at Anegada, British Virgin Islands. Natural Hazards 63, no. 1, 51-84.

Ball, D.F., 1964. Loss-on-ignition as an estimate of organic matter and organic carbon in non- calcareous soils. Journal of Soil Science 15, 84–92.

Barrientos, S. E., 1995. Dual seismogenic behavior: the 1985 Central Chile earthquake. Geophysical Research Letters, 22(24), 3541-3544.

Barrientos, S. E., 1996. On Predicting Coastal Uplift and Subsidence Due to Large Earthquakes in Chile. Third ISAG, ORSTOM, St Malo, France, 145-148.

Barrientos, S.E., 1997. Central Chile: an example of quasi-static crustal behavior. The Island Arc 6, 281-287.

Barrientos, S.E., 2007. Earthquakes in Chile, in: Moreno, T., Gibbons, W. (Eds.), Geology of Chile. The Geological Society, London, pp. 263-287.

Bartsch-Winkler, S., Schmoll, H.R., 1987. Earthquake-caused sedimentary couplets in the upper Cook Inlet region. U.S. Geological Survey, Circular 998, 92}95.

Bartsch-Winkler, S., Schmoll, H., 1993. Evidence for late Holocene relative sea-level fall from reconnaissance stratigraphical studies in an area of earthquake-subsided intertidal deposits, Isla Chiloe, southern Chile. In: Frostwick, L. E., Steel, R. J. (eds.) Tectonic controls and signatures in sedimentary successions. International Association of Seismologists. pp. 91- 108.

Beck, S., Barrientos, S.E., Kausel, E., Reyes, M., 1998. Source characteristics of historical earthquakes along central Chile subduction zone. Journal of South American Earth Sciences 11, 115-129.

Benson, B.E., Grimm, K.A., and Clague, J.J., 1997, Tsunami deposits beneath tidal marshes on northwestern Vancouver Island, British Colombia. Quaternary Research, v. 48, 192-204.

Berryman, K., Cooper, A., Norris, R., Villamor, P., Sutherland, R., Wright, T., Schermer, E.,Langridge, R., and Biasi, G., 2012. Late Holocene rupture history of the Alpine fault in South Westland, New Zealand. Bulletin of the Seismological Society of America,102(2), 620- 638., doi: 10.1785/0120110177.

Bilham, R., Engdahl, E.R., Feldl, N., Satyabala, S.P., 2005. Partial and complete rupture of the Indo-Andaman plate boundary 1847-2004. Seismological Research Letters, v. 76, p. 299-311.

Birks, H. J. B., 1986. Numerical zonation, comparison and correlation of Quaternary pollen- stratigraphical data. In: Handbook of Holocene Palaeoecology and Palaeohydrology. (Ed.: B. E. Berglund) John Wiley & Sons Ltd, London, 743-773.

208

Birks, H. J. B., 1992. Some reflections on the application of numerical methods in Quaternary palaeoecology. University of Joensuu, Publication Karelian Institute, 102, 7-20.

Birks, H.J.B., 1995. Statistical modelling of Quaternary science data. In: Maddy, D., Brew, J.S. (Eds.), Quantitative Paleoenvironmental Reconstructions. Quaternary Research Association, Cambridge 5, 161-254.

Blott, S.J., Pye, K., 2001. Gradistat: a grain size distribution and statistics package for the analysis of unconsolidated sediments. Earth Surface Processes and Landforms 26, 1237–1248.

Bondevik, S., Svendsen, J.I., Mangerud, J., 1997. Tsunami sedimentary facies deposited by the Storegga tsunami in shallow marine basins and coastal lakes, western Norway. Sedimentology 44, 1115–1131.

Bondevik, S., 2003. Storegga tsunami sand in peat below the Tapes beach ridge at Haroy, western Norway, and its possible relation to an early Stone Age settmement. Boreas 32(3), 476-483.

Bondevik, S., Mangerud, J., Dawson, S., Dawson, A., Lohne, O., 2005. Evidence for three North Sea tsunamis at the Shetland Islands between 8000 and 1500 years ago. Quaternary Science Reviews 24, 1757–1775.

Bourgeois, J., & Reinhart, M. A., 1989. Onshore erosion and deposition by the 1960 tsunami at the Rio Lingue estuary, south-central Chile. Eos Trans. AGU, 70(43), 1331.

Bourgeois, J., Petroff, C., Yeh, H., Titov, V., Synolakis, C.E., Benson, B., Kuroiwa, J., Lander, J., Norabuena, E., 1999. Geologic setting, field survey and modeling of the Chimbote, northern Peru, tsunami of 21 February 1996. Pure and Applied Geophysics 154, 513–540.

Bourgeois, J. (2006). Earthquakes: a movement in four parts? Nature, 440(7083), 430-431.

Bourgeois, J., 2009. Geologic effects and records of tsunamis, in: Robinson, A.R., Bernard, E.N. (Eds.), The Sea. Tsunamis, vol. 15. Harvard University Press, Cambridge, pp. 55–91.

Briggs, R. W., Engelhart, S. E., Nelson, A. R., Dura, T., Kemp, A. C., Haeussler, P. J., Corbett, D.R., Angster, S.J., and Bradley, L. A., 2014. Uplift and subsidence reveal a non‐persistent megathrust rupture boundary (Sitkinak Island, Alaska). Geophysical Research Letters. 41.7: 2289-2296.

Brill, D., Jankaew, K., Neubauer, N. P., Kelletat, D., Scheffers, A., Vött, A., & Brückner, H., 2014. Holocene coastal evolution of southwest Thailand–implications for the site-specific preservation of palaeotsunami deposits. Zeitschrift für Geomorphologie 58.1.

Bronk Ramsey, C., 2008, Depositional models for chronological records. Quaternary Science

209

Reviews, v. 27, no. 1-2, p. 42-60, doi: 10.1016/j.quascirev.2007.01.019.

Bronk Ramsey, C., 2009, Bayesian analysis of radiocarbon ages. Radiocarbon 51, no. 1, 337-360.

Caviedes, C., 1972. Geomorfologia del Cuaternario del Valle del Aconcagua, Chile central. Frieburger Geographische Hefte N. 11. 153 pp.

Chagué-Goff, C., Dawson, S., Goff, J. R., Zachariasen, J., Berryman, K. R., Garnett, D. L., Waldron, H. M., and Mildenhall, D. C., 2002. A tsunami (ca. 6300 years BP) and other Holocene environmental changes, northern Hawke's Bay, New Zealand. Sedimentary Geology, 150(1), 89-102.

Chagué-Goff, C., Schneider, J. L., Goff, J. R., Dominey-Howes, D., & Strotz, L., 2011. Expanding the proxy toolkit to help identify past events—lessons from the 2004 Indian Ocean Tsunami and the 2009 South Pacific Tsunami. Earth-Science Reviews, 107(1), 107-122.

Charles, D., 2002. Protocols for the analysis of algal samples collected as part of the U.S. Geological Survey National Water-Quality Assessment Program. Patrick Center for Environmental Research, Philadelphia: Academy of Natural Sciences, 2002.

Cifuentes, I., 1989. The 1960 Chilean earthquakes. Journal of Geophysical Research, v. 94, p. 665-680.

Cisternas, M., 2012. El terremoto de 1647 de Chile central como un evento intraplaca: ¿otra amenaza para Chile metropolitano? Revista de Geografía Norte Grande, 53:23-33.

Cisternas, M., Atwater, B. F., Torrejón, F., Sawai, Y., Machuca, G., Lagos, M., and Husni, M., 2005. Predecessors of the giant 1960 Chile earthquake. Nature 437(7057), 404-407.

Cisternas, M., Torrejón, F., and Gorigoitia, N., 2012. Amending and complicating Chile’s seismic catalog with the Santiago earthquake of 7 August 1580. Journal of South American Earth Sciences 33(1), 102–109.

Clague, J.J. and Bobrowsky, P.T., 1994. Evidence for a large earthquake and tsunami 100-400 years ago on western Vancouver Island, British Columbia. Quaternary Research, 41, 176- 184. Clague, J.J., Hutchinson, I., Mathewes, R.W., Patterson, R.T., 1999. Evidence for Late Holocene tsunamis at Catala Lake, British Columbia. Journal of Coastal Research 15(1), 45-60.

Cleve, P.T., 1894. Synopsis of the naviculoid diatom I and II. Kongliga Svenska Vetenskaps- Akademien 26, 1–194.

Cleve, P.T., 1895. Synopsis of the naviculoid diatom I and II. Kongliga Svenska Vetenskaps- Akademien 27, 1–129.

210

Cleve-Euler, A., 1922. Om diatomacevegetationen och dess forandringar i Sabysjon, Uppland samt negra damda sjoar i Salatrakten. Sveriges Geologiska Undersokning C309, 1–76.

Cochran, U., Berryman, K., Zachariasen, J., Mildenhall, D., Hayward, B., Southall, K., Hollis, K., and Wilson, K., 2006. Paleoecological insights into subduction zone earthquake occurrence, eastern North Island, New Zealand. Geological Society of America Bulletin, 118(9-10), 1051- 1074.

Cochran, U., Hannah, M., Harper, M., Van Dissen, R., Berryman, K., & Begg, J., 2007. Detection of large, Holocene earthquakes using diatom analysis of coastal sedimentary sequences, Wellington, New Zealand. Quaternary Science Reviews, 26(7), 1129-1147.

Combellick, R. A., 1986. Chronology of late‐Holocene earthquakes in southcentral Alaska: Evidence from buried organic soils in upper Turnagain Arm, Geol. Soc. Am. Abstr. Programs, 18(6), 569.

Combellick, R.A., 1991. Paleoseismicity of the Cook Inlet region, Alaska: Evidence from peat stratigraphy in Turnagain and Knik Arms. Alaska Division of Geological and Geophysical Surveys Professional Report, Fairbanks, 52 pp.

Combellick, R. A., 1994. Investigation of peat stratigraphy in tidal marshes along Cook Inlet, Alaska, to determine the frequency of 1964-style great earthquakes in the Anchorage region. Rep. of investigations/Alaska. Div. of Geol. and Geophys. Surveys, 94, 7.

Combellick, R., Reger, R.D., 1989. Evaluation of Holocene subsidence events of Cook Inlet estuarine flats near Anchorage, Alaska, as a basis for assessing seismic hazards in south- central Alaska. USGS Open-File Report 89-453, 445-446.

Combellick, R. A., & Reger, R. D., 1994. Sedimentological and radiocarbon-age data for tidal marshes along eastern and upper Cook Inlet, Alaska. State of Alaska, Department of Natural Resources, Division of Geological and Geophysical Surveys.

Comte, D., Eisemberg, A., Lorca, E., Pardo, M., Ponce, L., Saragoni, R., Singh, S.K., Suárez, G., 1986. The 1985 central Chile earthquake: a repeat of previous great earthquakes in the region? Science 233, 449-453.

Cooper, S., Gaiser, E.E., Wachnicka, A., 2010. Estuarine paleoecological reconstructions using diatoms. In: Smol, J.P., Stoermer, E.F. (Eds.), The Diatoms: Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge, pp. 324-345.

Darwin, C., 1839, Journal and Remarks. 1832-1836, in Narrative of the Surveying Voyages of His Majesty’s Ships Adventure and Beagle, Between the Years 1826 and 1836, v. III: London, Henry Colburn (reprinted in 1966 by AMS Press, Inc., New York).

Dawson, A., Shi, S., 2000. Tsunami deposits. Pure and Applied Geophysics 157, 875–897.

211

Dawson, S., and Smith, D.E, 2000. The sedimentology of middle Holocene tsunami facies in northern Sutherland, Scotland, UK. Marine Geology 170, no. 1–2, 69–79. doi:10.1016/S0025-3227(00)00066-9.

Dawson, S., 2007. Diatom biostratigraphy of tsunami deposits: examples from the 1998 Papua New Guinea tsunami. Sedimentary Geology 200, 328–335.

Darienzo, M. E., & Peterson, C. D., 1990. Episodic tectonic subsidence of late Holocene salt marshes, northern Oregon central Cascadia margin. Tectonics, 9(1), 1-22.

Darienzo, M.E., Peterson, C.D., Clough, C., 1994. Stratigraphic evidence for great subduction zone earthquakes at four estuaries in northern Oregon, USA. Journal of Coastal Research 10, 850-876.

Dawson, S., Smith, D.E., Ruffman, A., Shi, S., 1996a. The diatom biostratigraphy of tsunami deposits: Examples from recent and middle Holocene events. Physics and Chemistry of the Earth 21, 87-92.

Dawson, A.G., Shi, S., Dawson, S., Takahashi, T., Shuto, N., 1996b. Coastal sedimentation associated with the June 2nd and 3rd, 1994 tsunami in Rajegwesi, Java. Quaternary Science Reviews 15, 901-912.

Dawson, S. and Smith, D.E., 2000. The sedimentology of mid-Holocene tsunami facies in northern Scotland. Marine Geology 170, 69-79.

Dawson, S., 2007. Diatom biostratigraphy of tsunami deposits: examples from the 1998 Papua New Guinea tsunami. Sedimentary Geology 200, 328–335.

Dawson, A. G., & Stewart, I., 2007. Tsunami deposits in the geological record. Sedimentary Geology, 200(3), 166-183.

Denys, L., 1991. A check-list of the diatoms in the Holocene deposits of the western Belgian coastal plain with a survey of their apparent ecological requirements. Ministère des affaires économiques, Service Géologique de Belgique.

Dominey-Howes, D., Cundy, A., Croudace, I., 2000. High energy marine flood deposits on Astypalaea Island, Greece: possible evidence for the AD 1956 southern Aegean tsunami. Marine Geology 163, 303-315.

Dominey-Howes, Dale T.M., Geoff S. Humphreys, and Hesse, P.P., 2006. Tsunami and palaeotsunami depositional signatures and their potential value in understanding the late- Holocene tsunami record. The Holocene 16, no. 8, 1095–1107. doi:10.1177/0959683606069400.

212

Donato, S.V., Reinhardt EG, Boyce JI, Pilarczyk JE, Jupp BP, 2009. Particle-size distribution of inferred tsunami deposits in Sur Lagoon, Sultanate of Oman. Marine Geology, 257:54–64.

Dura, T., Rubin, C. M., Kelsey, H. M., Horton, B. P., Hawkes, A., Vane, C. H., Daryono, M., Grand Pre, C., Ladinsky, and T., Bradley, S., 2011. Stratigraphic record of Holocene coseismic subsidence, Padang, West Sumatra. Journal of Geophysical Research: Solid Earth,116(B11).

Dura, T., Cisternas, C., Horton, B.P., Ely, L.L., Nelson, A.R., Wesson, R.L., Pilarczyk, J.E., 2014. Coastal evidence for Holocene subduction-zone earthquakes and tsunamis in central Chile. Quaternary Science Reviews.

DuRoss, C.B., Personius, S.F., Crone, A.J., Olig, S.S., and Lund, W.R., 2011. Integration of paleoseismic data from multiple sites to develop an objective earthquake chronology: application to the Weber segment of the Wasatch fault zone, Utah. Bulletin of the Seismological Society of America, v. 101, no. 6, p. 2765–2781, doi: 10.1785/0120110102.

Ely, L.L., Cisternas, M., Wesson, R., Dura, T., 2014. Five centuries of tsunamis and land-level changes in the overlapping rupture area of the 1960 and 2010 Chilean earthquakes. Geology, G35830-1 Doi: 10.1130/G35830.1

Encinas, A., Herve, F., Villa Martinez, R., Nielsen, S., Finger, K., Peterson, D., 2006. Finding of a Holocene marine layer in Algarrobo (33°22'S), central Chile. Implications for coastal uplift Revista Geológica de Chile, Vol. 33, No. 2, p. 339-345.

Engelhart, S.E., Horton, B.P., Nelson, A.R., Hawkes, A.D., Witter, R.C., Wang, K., Wang, P-.L., Vane, C.H., 2013. Testing the use of microfossils to reconstruct great earthquakes at Cascadia. Geology doi, 10.1130/G34544.1

Farías, M., Vargas, G., Tassara, A., Carretier, S., Baize, S., Melnick, D., and Bataille, K., 2010. Land- level changes produced by the Mw 8.8 2010 Chilean earthquake. Science, v. 329, p. 916.

Fedorov, A. V., Brierley, C. M., and Emanuel, K., 2010. Tropical cyclones and permanent El Niño in the early Pliocene epoch. Nature, 463(7284), 1066-1070.

Fritz, H. M., Petroff, C. M., Catalán, P. A., Cienfuegos, R., Winckler, P., Kalligeris, N., and Synolakis, C. E., 2011. Field survey of the 27 February 2010 Chile tsunami. Pure and Applied Geophysics, 168(11), 1989-2010. doi:10.1007/s00024-011-0283-5.

Garrett, E., Shennan, I., Watcham, E.P., and Woodroffe, S.A., 2013. Reconstructing paleoseismic deformation, 1: Modern analogues from the 1960 and 2010 Chilean great earthquakes. Quaternary Science Reviews 75, no. 0: 11–21. doi:10.1016/j.quascirev.2013.04.007.

Garrett, E., Shennan, I., Woodroffe, S.A., Cisternas, M., Hocking, E.P., and Gulliver, P., 2014. Reconstructing paleoseismic deformation, 2: 1000 years of great earthquakes at Chucalen,

213

south central Chile. Quaternary Science Reviews 75, no. 0: 1-11. doi:10.1016/j.quascirev.2014.10.010.

Gelfenbaum, G., and Jaffe, B., 2003. Erosion and sedimentation from the 17 July 1998 Papua New Guinea tsunami. Pure and Applied Geophysics 160, 1969-1999.

Geller, R. J., 2011. Shake-up time for Japanese seismology. Nature 472, no. 7344: 407–409. doi:10.1038/nature10105.

Goff, J., Liu, P. L., Higman, B., Morton, R., Jaffe, B. E., Fernando, H., ... & Fernando, S., 2006. Sri Lanka field survey after the December 2004 Indian Ocean tsunami. Earthquake Spectra, 22(S3), 155-172.

Goff, J., Lamarche, G., Pelletier, B., Chagué-Goff, C., Strotz, L., 2011. Predecessors to the 2009 South Pacific tsunami in the Wallis and Futuna archipelago. Earth-Science Reviews 107(1-2), 91-106.

Goff, J., Chagué-Goff, C., Nichol, S., Jaffe, B., & Dominey-Howes, D., 2012. Progress in palaeotsunami research. Sedimentary Geology, 243, 70-88.

Goto, K., Chagué-Goff, C., Fujino, S., Goff, J., Jaffe, B., Nishimura, Y., Richmond, B., Sugawara, D., Szczuciński, W., Tappin, D.R., Witter, R., Yulianto, E., 2011. New insights of tsunami hazard from the 2011 Tohoku-oki event. Marine Geology 290, 46–50.

Grand Pre, C.A., Horton, B.P., Kelsey, H.M., Rubin, C.M., Hawkes, A.D., Daryono, M.R., Rosenberg, G., and Culver, S.J., 2012. Stratigraphic evidence for an early Holocene earthquake in Aceh, Indonesia. Quaternary Science Reviews 54, 142-151.

Grauert, M., Björck, S., and Bondevik, S., 2001, Storegga tsunami deposits in a coastal lake on Suduroy, the Faroe Islands. Boreas, v. 30, p. 263-271.

Guilbault, J.P., Clague, J.J., Lapointe, M., 1995. Amount of subsidence during a late Holocene earthquake – Evidence from fossil tidal marsh foraminifera at Vancouver Island, west coast of Canada. Palaeogeography Palaeoclimatology Palaeoecology 118, 49-71.

Guilbault, J.-P., Clague, J.J., and Lapointe, M., 1996, Foraminiferal evidene for the amount of coseismic subsidence during a late Holocene earhquake on Vancouver Island, west coast of Canada. Quaternary Science Reviews, v. 15, p. 913-937.

Hamilton, S., Shennan, I., Combellick, R., Mulholland, J., & Noble, C., 2005. Evidence for two great earthquakes at Anchorage, Alaska and implications for multiple great earthquakes through the Holocene. Quaternary Science Reviews, 24(18), 2050-2068.

214

Hanks, T.C., Beroza, G.C., and Toda, S., 2012, Have Recent Earthquakes Exposed Flaws in or Misunderstandings of Probabilistic Seismic Hazard Analysis? Seismological Research Letters, 83(5), 759–764, doi:10.1785/0220120043.

Hartley, B., Ross, R., and Williams, D.M., 1986. A check-list of the freshwater, brackish and marine diatoms of the British Isles and adjoining coastal waters. Journal of the Marine Biological Association of the United Kingdom 66, no. 03: 531-610.

Hawkes, A. D., Bird, M., Cowie, S., Grundy-Warr, C., Horton, B. P., Shau Hwai, A. T., and Aik, L. W., 2007. Sediments deposited by the 2004 Indian Ocean tsunami along the Malaysia– Thailand Peninsula. Marine Geology, 242 (1), 169-190.

Hawkes, A.D., Horton, B.P., Nelson, A.R., Vane, C.H., and Sawai, Y., 2010. The application of intertidal foraminifera to reconstruct coastal subsidence during the giant Cascadia earthquake of AD 1700 in Oregon, USA, Quaternary International, 122, 116-140.

Hawkes, A., Horton, B.P., Nelson, A., Vane, C., Sawai, Y., 2011. Coastal subsidence in Oregon, USA, during the giant Cascadia earthquake of AD 1700. Quaternary Science Reviews, 30, 364–376.

Hayward, B. W., Cochran, U., Southall, K., Wiggins, E., Grenfell, H. R., Sabaa, A., Shane, P.R., Gehrels, R. (2004). Micropalaeontological evidence for the Holocene earthquake history of the eastern Bay of Plenty, New Zealand, and a new index for determining the land elevation record. Quaternary Science Reviews, 23(14), 1651-1667.

Hayward, B. W., Grenfell, H. R., Sabaa, A. T., Carter, R., Cochran, U., Lipps, J. H., Shane, P.R., Morley, M. S., 2006. Micropaleontological evidence of large earthquakes in the past 7200 years in southern Hawke's Bay, New Zealand. Quaternary Science Reviews, 25(11), 1186- 1207.

Heki, K., 2011. A tale of two earthquakes. Science 332, no. 6036: 1390–1391. doi:10.1126/science.1206643.

Hemphill-Haley, E., 1995a. Diatom evidence for earthquake-induced subsidence and tsunami 300 years ago in southern coastal Washington. Geological Society of America Bulletin 107, 367-378.

Hemphill-Haley, E., 1995b. Intertidal diatoms from Willapa Bay, Washington: applications to studies of small-scale sea-level changes. Northwest Science, 69, 29–45.

Hemphill-Haley, E., 1996. Diatoms as an aid in identifying late-Holocene tsunami deposits. The Holocene, 6(4), 439-448.

215

Hogg, A. G., Hua, Q., Blackwell, P.G., Niu, M., Buck, C.E., Guilderson, T.P., Heaton, T.J., 2013. SHCAL13 Southern Hemisphere calibration, 0-50,000 Years CAL BP. Radiocarbon 55, no. 4 (2013).

Hong, I., 2014. Sedimentological and stratigraphic evidence for paleotsunami events at Quidico, Chile. Central Univerisity commons.

Horton, B. P., 1999. The distribution of contemporary intertidal foraminifera at Cowpen Marsh, Tees Estuary, UK: implications for studies of Holocene sea-level changes. Palaeogeography, Palaeoclimatology, Palaeoecology, 149(1), 127-149.

Horton, B.P., Edwards, R.J., 2005. The application of local and regional transfer functions to reconstruct former sea levels, North Norfolk, England. The Holocene 15, 216-228.

Horton, B.P., Edwards, B.P., 2006. Quantifying Holocene Sea-Level Change using Intertidal Foraminifera: Lessons from the British Isles. Cushman Foundation for Foraminiferal Research, Special Publication, vol. 40, 97 pp.

Horton, B. P., Zong, Y., Hillier, C., & Engelhart, S. (2007). Diatoms from Indonesian mangroves and their suitability as sea-level indicators for tropical environments. Marine Micropaleontology, 63(3), 155-168.

Horton, B.P., Rossi, V., Hawkes, A.D., 2009. The sedimentary record of the 2005 hurricane season from the Mississippi and Alabama coastlines. Quaternary International 195, 15-30.

Horton, B. P., Sawai, Y., Smol, J. P., & Stoermer, E. F., 2010. Diatoms as indicators of former sea levels, earthquakes, tsunamis, and hurricanes. The diatoms: applications for the environmental and earth sciences, (Ed. 2), 357-372.

Horton, B.P., Sawai, Y., Hawkes, A.D., Witter, R.C., 2011. Sedimentology and paleontology of a tsunami deposit accompanying the great Chilean earthquake of February 2010. Marine Micropaleontology, 79, 132–138.

Hughes, J.F., Mathewes, R.W., Clague, J.J., 2002. Use of pollen and vascular plants to estimate coseismic subsidence at a tidal marsh near Tofino, British Columbia. Palaeogeography Palaeoclimatology Palaeoecology 185, 145-161.

Hustedt, F., 1937. Systematische und okologische Untersuchungen der Diatomeen floren von Java, Bali, Sumatra. Archiv Hydrobiologie Supplement 15, 131–506.

Hustedt, F., 1939. Systematische und okologische Untersuchungen der Diatomeen floren von Java, Bali, Sumatra. Archiv Hydrobiologie Supplement 16, 1–394.

Hustedt, F., 1953. Diatomeen ausdem Naturschutzpark Seeon. Archiv f¨ur Hydrobiologie 47, 625–35.

216

Hustedt, F., 1957. Die Diatomeenflora des Fluss-systems der Weser im Gebiet der Hansestadt Bremen. Abhandlungen vom Naturwissenschaftlichen Verein zu Bremen 34, 181–440.

Hustedt, F., 1958. Diatomeen aus der Antarktis and dem Siidatlantik. Deutsche Antarkt. Exped. 1938/39, Wiss. Ergebn. 2(3): 103-191.

Hutchinson, I., Clague, J.J., and Mathewes, R.W., 1997, Reconstructing the tsumani record on an emerging coast: A case study of Kanim Lake, Vancouver Island, British Columbia, Canada: Journal of Coastal Research, v. 13, p. 545–553.

Jaffe, B., Buckley, M., Richmond, B., Strotz, L., Etienne, S., Clark, K., Watt, S., Gelfenbaum, G., and Goff, J., 2011. Flow speed estimated by inverse modeling of sandy sediment deposited by the 29 September 2009 tsunami near Satitoa, east Upolu, Samoa. Earth-Science Reviews 107, 23–37.

Jankaew, K., Atwater, B., Sawai, Y., Choowong, M., Charoentitirat, T., Martin, M., Prendergast, A., 2008. Medieval forewarning of the 2004 Indian Ocean tsunami in Thailand. Nature. 455, 1228-1231.

Jones, V.J., 2007. Diatom Introduction, in: Elias, S., (Ed.) Encyclopedia of Quaternary Science. Elsevier, Oxford, pp. 476 - 484.

Juggins, S., 1992, Diatoms in the Thames estuary, England: ecology, palaeoecology, and salinity transfer function. Bibliotheca Diatomologica 25, 216.

Kamatani, A., 1982. Dissolution rates of silica from diatoms decomposing at various temperatures. Marine Biology 68, 91–96.

Kelsey, H. M., Witter, R.C., and Hemphill-Haley, E., 2002. Plate-boundary earthquakes and tsunamis of the past 5500 yr, Sixes River estuary, southern Oregon. Geological Society of America Bulletin 114, no. 3: 298-314.

Kelsey, H.M., Nelson, A.R., Hemphill-Haley, E., and Witter, R.C., 2005, Tsunami history of an Oregon coastal lake reveals a 4600 yr record of great earthquakes on the Cascadia subduction zone. Geological Society of America Bulletin, v.117, no.7-8, pp.1009-1032.

Kelleher, J., 1972. Rupture zones of large South American earthquakes and some predictions. Journal of Geophysical Research 77, 2087-2103.

Kemp, A.C., Horton, B.P., Culver, S.J., Corbett, D.R., van de Plassche, O., Gehrels, W.R., Douglas, B.C., Parnell, A.C., 2009. Timing and magnitude of recent accelerated sea-level rise (North Carolina, United States). Geology 37, 1035-1038.

217

Kemp A.C., Nelson A.R., and Horton B.P., 2013. Radiocarbon Dating of Plant Macrofossils from Tidal-Marsh Sediment. In: John F. Shroder (ed.) Treatise on Geomorphology, Volume 14, pp. 370-388. San Diego: Academic Press.

Kortekaas, S., Dawson, A. G., 2007. Distinguishing tsunami and storm deposits: an example from Martinhal, SW Portugal. Sedimentary Geology 200, 208-221.

Krammer, K., Lange-Bertalot, H., 1986. Bacillarophyceae 2/1. Naviculaceae, in: Ettl, H., Gerloff, J., Heynig, H., Mollenhauser (Eds.), Süsswasserflora von Mitteleuropa. Gustav Fischer Verlag, Stuttgart, pp. 1–876.

Krammer, K., Lange-Bertalot, H., 1988. Bacillarophyceae 2/2. Basillariaceae, Epithemiaceae, Surirellaceae, in: Ettl, H., Gerloff, J., Heynig, H., Mollenhauser (Eds.), Süsswasserflora von Mitteleuropa. Gustav Fischer Verlag, Stuttgart, pp. 1–600.

Krammer, K., Lange-Bertalot, H., 1991a. Bacillarophyceae 2/3. Centrales, Fragilariaceae, Eunotiaceae, in: Ettl, H., Gerloff, J., Heynig, H., Mollenhauser (Eds.), Süsswasserflora von Mitteleuropa. Gustav Fischer Verlag, Stuttgart, pp. 1–600.

Krammer, K., Lange-Bertalot, H., 1991b. Bacillarophyceae 2/4. Achnanthaceae, Kritische Erganzungen zu Navicula (Lineolatae) und Gomphonema, in: Ettl, H. (Ed.), Pascher's Süsswasserflora von Mitteleuropa, vol. 2, part 4. Gustav Fischer Verlag, Stuttgart, pp. 1–437.

Lange-Bertalot, H., Iconographia Diatomologica Annotated Diatom Micrographs – Diatom Flora of Marine Coasts I, Edited: H. Lange-Bertalot; Vol. 7 Pub: Koeltz Scientific Books, 2000. Copyright: A.R.G. Gantner Verlag K.G. By: Andrzej Witkowiski, Horst Lange- Bertalot and Ditmar Metzeltin

Lay, T., Ammon, C. J., Kanamori, H., Koper, K. D., Sufri, O., & Hutko, A. R., 2010. Teleseismic inversion for rupture process of the 27 February 2010 Chile (Mw 8.8) earthquake. Geophysical Research Letters, 37(13).

Leonard, L.J., Currie, C.A., Mazzotti, S., Hyndman, R.D., 2010. Rupture area and displacement of past Cascadia great earthquakes from coastal coseismic subsidence. Geological Society of America Bulletin 112, 2079-2096.

Leonard, L.J., Hyndman, R.D., Mazzotti, S., 2004. Coseismic subsidence in the 1700 great Cascadia earthquake: Coastal estimates versus elastic dislocation models. Geological Society of America Bulletin, 116: 655-670.

Lepland, A., Miller, U., & Sakson, M., 1995. Palaeoenvironmental conditions during the Baltic Yoldia stage in the Tallinn area, northern Estonia. Quaternary International, 27, 83-94.

Lomnitz, C., 1970. Major earthquakes and tsunamis in Chile during the period 1535 to 1955. International Journal of Earth Sciences 59, 938-960.

218

Lomnitz, C., 2004. Major earthquakes of Chile: a historical survey, 1535-1960. Seismological Research Letters 75, no. 3: 368-378.

Lorito, S., Romano, F., Atzori, S., Tong, X., Avallone, A., McCloskey, J., Cocco, M., Boschi, E., and Piatanesi, A., 2011. Limited overlap between the seismic gap and coseismic slip of the great 2010 Chile earthquake. Nature Geoscience, 4(3), 173-177.

Lowe, R. L., 1974. Environmental Requirements and Pollution Tolerance of Freshwater Diatoms. EPA-670/4-74-005. Cincinnati, OH. 344 pp.

Lu, H., & Liu, K. B., 2003. Phytoliths of common grasses in the coastal environments of southeastern USA. Estuarine, Coastal and Shelf Science,58(3), 587-600.

Maldonado, A., and Villagrán, C., 2006. Climate variability over the last 9900 cal yr BP from a swamp forest pollen record along the semiarid coast of Chile. Quaternary Research 66, no. 2: 246–258. doi:10.1016/j.yqres.2006.04.003.

Malik, J. N., Shishikura, M., Echigo, T., Ikeda, Y., Satake, K., Kayanne, H., Sawai, Y., Murty, C.V.R., and Dikshit, O., 2011. Geologic evidence for two pre-2004 earthquakes during recent centuries near Port Blair, South Andaman Island, India. Geology, 39(6), 559-562.

Mamo, B., Strotz, L., and Dominey-Howes, D., 2009. Tsunami sediments and their foraminiferal assemblages. Earth-Science Reviews 96, no. 4: 263-278.

Martínez, C., Quezada, M., & Rubio, P., 2011. Historical changes in the shoreline and littoral processes on a headland bay beach in central Chile. Geomorphology, 135(1), 80-96.

May, S. M., Pint, A., Rixhon, G., Kelletat, D., Wennrich, V., and Brückner, H., 2013. Holocene coastal stratigraphy, coastal changes and potential palaeoseismological implications inferred from geo-archives in Central Chile (29–32° S). Zeitschrift für Geomorphologie, Supplementary Issues, 57(4), 201-228.

Melnick, D., Bookhagen, B., Strecker, M. R., and Echtler, H. P., 2009. Segmentation of megathrust rupture zones from fore‐arc deformation patterns over hundreds to millions of years, Arauco peninsula, Chile. Journal of Geophysical Research: Solid Earth (1978– 2012), 114(B1).

Melnick, D., Cisternas, M., Moreno, M., and Norambuena, R., 2012. Estimating coseismic coastal uplift with an intertidal mussel: calibration for the 2010 Maule Chile earthquake (Mw= 8.8). Quaternary Science Reviews, 42, 29-42.

Métois, M., Socquet, A., and Vigny, C., 2012. Interseismic coupling, segmentation and mechanical behavior of the central Chile subduction zone. Journal of Geophysical Research: Solid Earth 117, no. B3: B03406. doi:10.1029/2011JB008736.

219

Milne, G. A., Long, A. J., and Bassett, S. E., 2005. Modeling Holocene relative sea‐level observations from the Caribbean and South America. Quaternary Science Reviews, 24(10– 11), 1183–1202, doi:10.1016/j.quascirev. 2004.10.005.

Minoura, K., Nakaya, S., 1991. Traces of tsunami preserved in inter-tidal lacustrine and marsh deposits – some examples from northeast Japan. Journal of Geology 99(2), 265-287.

Minoura, K., Imamura, F., Sugawara, D., Kono, Y., 2001. The 869 Jogan tsunami deposit and recurrence interval of large-scale tsunami on the Pacific coast of northeast Japan. Journal of Natural Disaster Science 23(2), 83-88.

Mitrovica, J., and Milne, G., 2002. On the origin of late Holocene sea‐level highstands within equatorial ocean basins. Quaternary Science Reviews 21(20–22), 2179–2190, doi:10.1016/S0277-3791(02)00080-X.

Moernaut, J., M. De Batist, F. Charlet, K. Heirman, E. Chapron, M. Pino, R. Brümmer, and R. Urrutia, 2007. Giant earthquakes in south-central Chile revealed by Holocene mass-wasting events in Lake Puyehue, Sedimentary Geology 195, 239–256.

Moernaut, J., De Batist, M., Heirman, K., Van Daele, M., Pino, M., Brümmer, R., & Urrutia, R., 2009. Fluidization of buried mass-wasting deposits in lake sediments and its relevance for paleoseismology: results from a reflection seismic study of lakes Villarrica and Calafquén (South-Central Chile). Sedimentary Geology, 213(3), 121-135.

Moernaut, J., Daele, M. V., Heirman, K., Fontijn, K., Strasser, M., Pino, M., Urrutia, R., and De Batist, M., 2014. Lacustrine turbidites as a tool for quantitative earthquake reconstruction: New evidence for a variable rupture mode in south central Chile. Journal of Geophysical Research: Solid Earth, 119(3), 1607-1633.

Monecke, K., Finger, W., Klarer, D., Kongko, W., McAdoo, B., Moore, A., and Sudrajat, S., 2008. A 1,000 year sediment record of tsunami recurrence in northern Sumatra. Nature, v. 455, p. 1232-1234.

Montessus de Ballore, F., 1912. Historia sísmica de los Andes Meridionales al sur del paralelo XVI. Segunda parte. Imprenta Cervantes, Santiago, 236 p.

Moreno, T., and Gibbons, W. (eds), 2007. The Geology of Chile. The Geological Society of London, 21-114.

Moreno, M. S., Bolte, J., Klotz, J., and Melnick, D., 2009. Impact of megathrust geometry on inversion of coseismic slip from geodetic data: Application to the 1960 Chile earthquake. Geophysical Research Letters, 36(16), L16310. doi:10.1029/2009GL039276

220

Moreno, M., Melnick, D., Rosenau, M., Bolte, J., Klotz, J., Echtler, H., ... & Oncken, O., 2011. Heterogeneous plate locking in the South–Central Chile subduction zone: Building up the next great earthquake. Earth and Planetary Science Letters, 305(3), 413-424.

Moreno, M., Melnick, D., Rosenau, M. [and 17 others], 2012, Toward understanding tectonic control on the Mw 8.8 2010 Maule Chile earthquake: Earth and Planetary Science Letters, v. 321-322, p. 152-165, doi:10.1016/j.epsl.2012.01.006.

Morton, R.A., Gelfenbaum, G., and Jaffe, B.E., 2007. Physical criteria for distinguishing sandy tsunami and storm deposits using modern examples. Sedimentary Geology 200, 184–207.

Munsell soil color charts. Macbeth division of Kollmorgen Corporation, 1975.

Nanayama, F., Satake, K., Furukawa, R., Shimokawa, K., Atwater, B.F., Shigeno, K., Yamaki, S., 2003. Unusually large earthquakes inferred from tsunami deposits along the Kuril trench. Nature 424, 660-663.

Nanayama, F., Furukawa, R., Shigeno, K., Makino, A., Soeda, Y., Igarashi, Y., 2007. Nine unusually large tsunami deposits from the past 4000 years at Kiritappu marsh along the southern Kuril Trench. Sedimentary Geology 200, 275–294.

Nelson, A.R., Kashima, K., 1993. Diatom zonation in southern Oregon tidal marshes relative to vascular plants, foraminifera, and sea level. Journal of Coastal Research 9, 673-697.

Nelson, A.R., Atwater, B.F., Bobrowsky, P.T., Bradley, L.A., Clague, J.J., Carver, G.A., Darienzo, M.E., Grant, W.C., Krueger, H.W., Sparks, R., Stafford, T.W., Stuiver, M., 1995. Radiocarbon evidence for extensive plate-boundary rupture about 300 years ago at the Cascadia subduction zone. Nature 378, 371-374.

Nelson, A.R., Jennings, A.E., and Kashima, K., 1996a. Holocene intertidal stratigraphy, microfossils, rapid submergence, and earthquake recurrence at Coos Bay, southern coastal Oregon, USA. Geological Society of America Bulletin, v. 108, p. 141-154.

Nelson, A.R., Shennan, I., and Long, A.J., 1996b. Identifying coseismic subsidence in tidal- wetland stratigraphic sequences at the Cascadia subduction zone of western North America. Journal of Geophysical Research 101, 6115-6135.

Nelson, A. R., Ota, Y., Umitsu, M., Kashima, K., & Matsushima, Y., 1998. Seismic or hydrodynamic control of rapid late-Holocene sea-level rises in southern coastal Oregon, USA? The Holocene, 8(3), 287-299.

Nelson, A. R., Kelsey, H. M., and Witter, R. C., 2006. Great earthquakes of variable magnitude at the Cascadia subduction zone. Quaternary Research, 65(3), 354-365.

221

Nelson, A. R., Sawai, Y., Jennings, A. E., Bradley, L. A., Gerson, L., Sherrod, B. L., Sabean, J., Horton, B. P., 2008. Great-earthquake paleogeodesy and tsunamis of the past 2000 years at Alsea Bay, central Oregon coast, USA. Quaternary Science Reviews, 27(7), 747-768.

Nelson, A.R., Kashima, K., and Bradley, L., 2009. Fragmentary evidence of great‐earthquake subsidence during Holocene emergence, Valdivia Estuary, south central Chile. Bulletin of the Seismological Society of America 99(1), 71–86.

Nichol, S. L., Goff, J. R., Devoy, R. J. N., Chagué-Goff, C., Hayward, B., & James, I. (2007). Lagoon subsidence and tsunami on the West Coast of New Zealand. Sedimentary Geology, 200(3), 248-262.

Nishimura, Y., and Miyaji, N., 1995. Tsunami deposits from the 1993 southwest Hokkaido earthquake and the 1640 Hokkaido Komagatake eruption, northern Japan. Pure and Applied Geophysics 144, no. 3-4: 719-733.

Ota, Y., Miyauchi, T., Paskoff, R., and Koba, M., 1995. Plio-Quaternary marine terraces and their deformation along the Altos de Talinay, north-central Chile. Andean Geology 22, no. 1: 89- 102.

Ovenshine, A. T., & Lawson, D. E., 1976. The Placer River Silt--An intertidal deposit caused by the 1964 Alaska earthquake.

Palmer, A.J.M., Abbott, W.H., 1986. Diatoms as indicators of sea-level change. In: can de Plassche, O. (Ed.), Sea-level Research: A Manual for the Collection and Evaluation of Data. Geobooks, Norwich, pp. 457-488.

Pararas-Carayannis, G., 2003. Near and far-field effects of tsunamis generated by the PAroxysmal eruptions, explosions, Caldera collapses and massive slope failures of the Krakatau volcano in Indonesia on August 26-27, 1883. Science of Tsunami Hazards.

Pardo, M., Comte, D. and Monfret, T., 2002. Seismotectonic and stress distribution in the central Chile subduction zone. Flat-Slab Subduction in the Andes 15, no. 1: 11–22. doi:10.1016/S0895-9811(02)00003-2.

Parsons, M.L., 1998. Salt marsh sedimentary record of the landfall of Hurricane Andrew on the Louisiana coast: diatoms and other paleoindicators. Journal of Coastal Research 14, 939– 950.

Patterson, R.T., Dalby, A.P., Roe, H.M., Guilbault, J.-P., Hutchinson, I., Clague, J.J., 2005. Relative utility of foraminifera, diatoms and macrophytes as high resolution indicators of paleo-sea level in coastal British Columbia, Canada. Quaternary Science Reviews 24, 2002-2014.

Patrick, R. M., and Reimer, C. W., 1966, The diatoms of the United States exclusive of Alaska and Hawaii, Volume 1: Monographs of the Academy of Natural Sciences of Philadelphia, 688.

222

Patrick, R. M., and Reimer, C. W., 1975, The diatoms of the United States exclusive of Alaska and Hawaii, Volume 2: Monographs of the Academy of Natural Sciences of Philadelphia, 213.

Patrick, R., Williams, D.D., 1990. Aquatic biota in North America. In: Wolman, M.G., Riggs, H.D., (eds.), Surface water hydrology: Boulder, Colorado, Geological Society of America, the Geology of North America, v 0-1.

Peltier, W.R., 2004. Global glacial isostasy and the surface of the ice-age Earth: The ICE-5G (VM2) model and GRACE. Annual Review of Earth and Planetary Sciences, 32, 111-149.

Peltier, W.R., Drummond, R., 2008. Rheological stratification of the lithosphere: a direct inference based upon the geodetically observed pattern of the glacial isostatic adjustment of the North American continent. Geophysical Research Letters 35. doi:10.1029/2008GL034586.

Peterson, C. D., Carver, G. A., Cruikshank, K. M., Abramson, H. F., Garrison‐Laney, C. E., & Dengler, L. A., 2011. Evaluation of the use of paleotsunami deposits to reconstruct inundation distance and runup heights associated with prehistoric inundation events, Crescent City, southern Cascadia margin. Earth Surface Processes and Landforms, 36(7), 967- 980.

Pilarczyk, J.E., Horton, B.P., Witter, R.C., Vane, C.H., Chagué-Goff, C., Goff, J., 2012. Sedimentary and foraminiferal evidence of the 2011 Tohoku-oki tsunami on the Sendai coastal plain, Japan. Sedimentary Geology 282, 78-89.

Pilarczyk, J. E., Dura, T., Horton, B. P., Engelhart, S. E., Kemp, A. C., & Sawai, Y., 2014. Microfossils from coastal environments as indicators of paleo-earthquakes, tsunamis and storms. Palaeogeography, Palaeoclimatology, Palaeoecology.

Plafker, G., 1965. Tectonic deformation associated with the 1964 Alaska earthquake. Science 148, 1675-1687.

Plafker, G., 1969. Tectonics of the March 27, 1964, Alaska earthquake. U.S. Geological Survey Professional Paper 543-1, 1-74.

Plafker, G., and Savage, J. C., 1970. Mechanism of the Chilean earthquakes of May 21 and 22, 1960. Geological Society of America Bulletin 81, 1001–-1030.

Plafker, G., Lajoie, K. R., & Rubin, M., 1992. Determining recurrence intervals of great subduction zone earthquakes in southern Alaska by radiocarbon dating. In Radiocarbon after Four Decades (pp. 436-453). Springer, New York.

Plater, A.J., Kirby, J.R., Boyle, J.F., Shaw, T., and Mills, H., 2015. Loss On Ignition And Organic Content, In Handbook Of Sea-Level Research.

223

Plant, N.G., Stockdon, H.F., Sallenger, A.H., Turco, M.J., East, J.W., Taylor, A.A., Shaffer, W.A., 2010. Forecasting Hurricane impact on coastal topography. Eos 91(7), 65-72.

Rebolledo, L., Lange, C. B., Figueroa, D., Pantoja, S., Muñoz, P., & Castro, R., 2005. 20th century fluctuations in the abundance of siliceous microorganisms preserved in the sediments of the Puyuhuapi channel (44 S), Chile. Revista Chilena de Historia Natural, 78, 469-488.

Rebolledo, L., González, H. E., Muñoz, P., Iriarte, J. L., Lange, C. B., Pantoja, S., & Salamanca, M., 2011. Siliceous productivity changes in Gulf of Ancud sediments (42° S, 72° W), southern Chile, over the last∼ 150 years. Continental Shelf Research, 31(3), 356-365.

Rhodes, B., Tuttle, M., Horton, B.P., Doner, L., Kelsey, H.M., Nelson, A., and Cisternas, M., 2006. Paleotsunami research. Eos, Transactions American Geophysical Union 87, no. 21: 205-209.

Rivera, R., 2000. Guide for references and distribution for the class Bacillariophyceae in Chile between 18⁰28'S and 58⁰S. J. Cramer.

Riznyk, R. Z., 1973, Interstitial diatoms from two tidal flats in Yaquina Estuary, Oregon USA: Botanica Marina, v. 16, p. 113–138.

Round, F. E., Crawford, R. M., & Mann, D. G., 1990. The diatoms: biology & morphology of the genera. Cambridge University Press.

Ruegg, J. C., Rudloff, A., Vigny, C., Madariaga, R., De Chabalier, J. B., Campos, J., and Dimitrov, D., 2009. Interseismic strain accumulation measured by GPS in the seismic gap between Constitución and Concepción in Chile. Physics of the Earth and Planetary Interiors, 175(1), 78-85.

Rühland, K., Smol, J. P., Jasinski, J. P. P., & Warner, B., 2000. Response of diatoms and other siliceous indicators to the developmental history of a peatland in the Tiksi Forest, Siberia, Russia. Arctic, Antarctic and Alpine Research 32, 167–78.

Sacchi, M., Insinga, D., Milia, A., Molisso, F., Raspini, A., Torrente, M. M., & Conforti, A., 2005. Stratigraphic signature of the Vesuvius 79 AD event off the Sarno prodelta system, Naples Bay. Marine geology, 222, 443-469.

Saillard, M., Hall, S. R., Audin, L., Farber, D. L., Hérail, G., Martinod, J., ... & Bondoux, F., 2009. Non-steady long-term uplift rates and Pleistocene marine terrace development along the Andean margin of Chile (31° S) inferred from 10Be dating. Earth and Planetary Science Letters, 277(1), 50-63.

Sambridge M., Braun, J., McQueen, H., 1995. Geophysical parametrization and interpolation of irregular data using natural neighbors. Geophysical Journal International 122, 837–857.

224

Satake, K. (2007) Volcanic origin of the 1741 Oshima-Oshima tsunami in the Japan Sea. Earth Planets Space, 59, 381-390.

Satake, K., and Atwater, B. F., 2007. Long-term perspectives on giant earthquakes and tsunamis at subduction zones. Annual Review of Earth and Planetary Science, 35, 349-374.

Sawai, Y., 2001a. Episodic emergence in the past 3000 years at the Akkeshi Estuary, Hokkaido, Northern Japan. Quaternary Research, 56, 231-241.

Sawai Y., 2001b. Distribution of living and dead diatoms in tidal wetlands of northern Japan : relations to taphonomy. Palaeogeography Palaeoclimatology Palaeoecology, 173, 125-141.

Sawai, Y., Nasu, H., Yasuda, Y., 2002. Fluctuations in relative sea-level during the past 3000 yr in the Onnetoh Estuary, Hokkaido, northern Japan. Journal of Quaternary Science, 17, 607-622.

Sawai, Y., Satake, K., Takanobu, K., Nasu, H., Shishikura, M., Atwater, B.F., Horton, B.P., Kelsey, H.M., Nagumo, T., and Yamaguchi, M., 2004a. Transient uplift after a 17th-century earthquake along the Kuril subduction zone. Science 306, 1918-1920.

Sawai, Y., Horton, B.P., Nagumo, T., 2004b. The development of a diatom-based transfer function along the Pacific coast of eastern Hokkaido, northern Japan – an aid in paleoseismic studies of the Kuril subduction zone. Quaternary Science Reviews 23, 2467-2483.

Sawai, Y., Fujii, Y., Fujiwara, O., Kamataki, T., Komatsubara, J., Okamura, Y., ... & Shishikura, M., 2008. Marine incursions of the past 1500 years and evidence of tsunamis at Suijin-numa, a coastal lake facing the Japan Trench.The Holocene, 18(4), 517-528.

Sawai, Y., Jankaew, K., Martin, M.E., Prendergast, A., Choowong, M., Charoentitirat, T., 2009a. Diatom assemblages in tsunami deposits associated with the 2004 Indian Ocean tsunami at Phra Thong Island, Thailand. Marine Micropaleontology, 73, 70-79.

Sawai, Y., Kamataki, T., Shishikura, M., Nasu, H., Okamura, Y., Satake, K., Thomson, K.H., Matsumoto, D., Fujii, Y., Komatsubara, J., Aung, T.T. (2009b): Aperiodic recurrence of geologically recorded tsunamis during the past 5500 years in eastern Hokkaido , Japan . Journal of Geophysical Research, 114, B01319, doi:10.1029/2007JB005503

Sawai, Y., Namegaya, Y., Okamura, Y., Satake, K., & Shishikura, M., 2012. Challenges of anticipating the 2011 Tohoku earthquake and tsunami using coastal geology. Geophysical Research Letters, 39(21)

Shennan, I., Long, A.J., Rutherford, M.M., Green, F.M., Innes, J.B., Lloyd, J.M., Zong, Y., Walker, K.J., 1996. Tidal marsh stratigraphy, sea level change and large earthquakes: I. A 5000 year record in Washington. Quaternary Science Reviews 15, 1023-1059.

225

Shennan, I., Long, A.J., Rutherford, M.M., Kirkby, J., Green, F.M., Innes, J.B., Walker, K., 1998. Tidal marsh stratigraphy, sea level change and large earthquakes, II: Events during the last 3500 years at Netarts Bay, Oregon, USA. Quaternary Science Reviews, 17, 365-393.

Shennan, I., Scott, D. B., Rutherford, M., Zong, Y. Q., 1999. Microfossil analysis of sediments representing the 1964 earthquake, exposed at Girdwood Flats, Alaska, USA. Quaternary International, 60, 55-73.

Shennan, I., & Hamilton, S., 2006. Coseismic and pre-seismic subsidence associated with great earthquakes in Alaska. Quaternary Science Reviews, 25(1), 1-8.

Shennan, I., Bruhn, R., & Plafker, G., 2009. Multi-segment earthquakes and tsunami potential of the Aleutian megathrust. Quaternary Science Reviews, 28(1), 7-13.

Shennan, I., Bruhn, R., Barlow, N., Good, K., & Hocking, E. (2014a). Late Holocene great earthquakes in the eastern part of the Aleutian megathrust. Quaternary Science Reviews, 84, 86-97.

Shennan, I., Barlow, N., Carver, G., Davies, F., Garrett, E., & Hocking, E. (2014b). Great tsunamigenic earthquakes during the past 1000 yr on the Alaska megathrust. Geology, 42(8), 687-690.

Sherrod, B. L. (1999). Gradient analysis of diatom assemblages in a Puget Sound salt marsh: can such assemblages be used for quantitative paleoecological reconstructions? Palaeogeography, Palaeoclimatology, Palaeoecology, 149(1), 213-226.

Sherrod, B. L., 2001. Evidence for earthquake-induced subsidence about 1100 yr ago in coastal marshes of southern Puget Sound, Washington. Geological Society of America Bulletin 113, no. 10: 1299-1311.

Shiki, T., Tachibana, T., Fujiwara, O., Goto, K., Nanayama, F., Yamazaki, T., 2008. Characteristic Features of Tsunamiites, in: Shiki, T., Tsuji, Y., Yamasaki, T., Minoura, K. (Eds.), Tsunamiites — Features and Implications. Elsevier, pp. 319–340.

Sieh, K., Natawidjaja, D.H., Meltzner, A.J., Shen, C., Cheng, H., Li, K., Suwargadi, B.W., Galetzka, J., Philibosian, B., and Edwards, R.L., Earthquake supercycles inferred from sea-level changes recorded in the corals of west Sumatra. Science 322, no. 5908 (2008): 1674-1678.

Simons, M., Minson, S.E., Sladen, A., Ortega, F., Jiang, J., Owen, S.E., Meng, L., Ampuero, J-P., Wei, S., Chu, R., Helmberger, D.V., Kanamori, H., Hetland, E., Moore, A.W., Webb, F.H. 2011. The 2011 magnitude 9.0 Tohoku-oki earthquake: mosaicking the megathrust from seconds to centuries. Science 332, 1421-1425.

Small, C., Gornitz, V., & Cohen, J. E., 2000. Coastal hazards and the global distribution of human population. Environmental Geosciences, 7(1), 3-12.

226

Smith, D.E., Shi, S., Cullingford, R.A., Dawson, A.G., Dawson, S., Firth, C.R., Foster, I.D.L., Fretwell, P., Haggart, B.A., Holloway, L.K., Long, D., 2004. The Holocene Storegga slide tsunami in the United Kingdom. Quaternary Science Reviews 23/24, 2291–2321.

Soloviev, S.L., Go, Ch.N., 1984. Catalogue of tsunamis on the eastern shore of the Pacific Ocean. Canadian Translation of Fisheries and Aquatic Sciences No. 5078, Canada Institute for Scientific and Technical Information, National Research Council, Ottawa, 293 pp.

Stein, S., and Okal, E.A., 2007. Ultralong period seismic study of the December 2004 Indian Ocean earthquake and implications for regional tectonics and subduction process. Bulletin of the Seismological Society of America 97, S279-S295.

Stein, S., & Okal, E. A., 2011. The size of the 2011 Tohoku earthquake need not have been a surprise. Eos, Transactions American Geophysical Union, 92 (27), 227-228.

Stewart, I. S., & Vita-Finzi, C. (Eds.). (1998). Coastal tectonics. Geological Society of London.

Subarya, C., Chlieh, M., Prawirodirdjo, L., Avouac, J. P., Bock, Y., Sieh, K., Meltzner, A.J., Natawidjaja, D.H., and McCaffrey, R. (2006). Plate-boundary deformation associated with the great Sumatra–Andaman earthquake. Nature, 440(7080), 46-51.

Switzer, A.D., and Jones, B.G., 2008. Large-scale washover sedimentation in a freshwater lagoon from the southeast Australian coast: sea-level change, tsunami or exceptionally large storm? The Holocene 18, no. 5: 787–803. doi:10.1177/0959683608089214.

Szczuciński, W., 2012. The post-depositional changes of the onshore 2004 tsunami deposits on the Andaman Sea coast of Thailand. Natural Hazards 60(1), 115-133.

Szczuciński, W., Kokociński, M., Rzeszewski, M., Chagué-Goff, C., Cachão, M., Goto, K., and Sugawara, D., 2012. Sediment sources and sedimentation processes of 2011 Tohoku-oki tsunami deposits on the Sendai Plain, Japan — Insights from diatoms, nannoliths and grain size distribution. Sedimentary Geology, (0), doi:10.1016/j.sedgeo.2012.07.019.

Szkornik, K., Gehrels, W. R., & Kirby, J. R., 2006. Using salt-marsh diatoms to reconstruct sea- level changes in Ho Bugt western Denmark. Marine Geology, 235, 137–50.

Telford, R. J., & Birks, H. J. B. (2011a). A novel method for assessing the statistical significance of quantitative reconstructions inferred from biotic assemblages. Quaternary Science Reviews, 30(9), 1272-1278.

Telford, R. J., & Birks, H. J. B., 2011. Effect of uneven sampling along an environmental gradient on transfer-function performance. Journal of Paleolimnology, 46(1), 99-106.

227

Tooley, M. J., 1978. Sea-level changes: northwest England during the Flandrian Stage. Clarendon Press, Oxford. 232 pp.

Tuttle, Martitia P., Ruffman, A., Anderson, T., and Jeter, H., 2004. Distinguishing tsunami from storm deposits in eastern North America: The 1929 Grand Banks tsunami versus the 1991 Halloween storm. Seismological Research Letters 75, no. 1: 117–131. doi:10.1785/gssrl.75.1.117.

Troels-Smith, J., 1955. Characterisation on unconsolidated sediments. Dan. Geol. Unders. Ser., IV, 3, 1-73.

Udías, A., Madariaga, R. Buforn, E., Muñoz, D,. and Ros, M., 2012. The large Chilean historical earthquakes of 1647, 1657, 1730, and 1751 from contemporary documents. Bulletin of the Seismological Society of America 102, no. 4: 1639–1653. doi:10.1785/0120110289. van de Plassche, O., Erkens, G., van Vliet, F., Brandsma, J., van der Borg, K., de Jong, A.F.M., 2006. Salt-marsh erosion associated with hurricane landfall in southern New England in the fifteenth and seventeenth centuries. Geology 34(10), 829-832. van der Werff, H., and Huls, H., 1958–1974. Diatomeeenflora van Nederland, 8 parts, published privately, De Hoef, The Netherlands.

Villa-Martínez, R., and Villagrán, C., 1997. Historia de La Vegetación de Bosques Pantanosos de La Costa de Chile Central Durante El Holoceno Medio Y Tardío. Revista Chilena de Historia Natural 70, no. 3: 391–401.

Vigny, C., Socquet, A., Peyrat, S., Ruegg, J.-C., Métois, M., Madariaga, R., Morvan, S., 2011. The 2010 Mw 8.8 Maule megathrust earthquake of central Chile, Monitored by GPS. Science 332, no. 6036: 1417–1421. doi:10.1126/science.1204132.

Vos, P . C ., 1986 . De sediment stabiliserende werking van benthische diatomeeen in het intergetijde gebied van de Oosterschelde . GEOMOR-nota 86-03, Rijkswaterstaat, Dienst Getijde-wateren, Middelburg, 156 pp .

Vos, P. C., and H. de Wolf, Methodological aspects of paleo-ecological diatom research in coastal areas of the Netherlands, Neth. J. Geosci./Geol. Mijnb., 1988, 67: 31–40.

Vos, Peter C., and de Wolf, H., 1993. Diatoms as a tool for reconstructing sedimentary environments in coastal wetlands; methodological aspects. In Twelfth International Diatom Symposium, pp. 285-296. Springer Netherlands.

Wang, P. L., Engelhart, S. E., Wang, K., Hawkes, A. D., Horton, B. P., Nelson, A. R., & Witter, R. C. (2013). Heterogeneous rupture in the great Cascadia earthquake of 1700 inferred from coastal subsidence estimates. Journal of Geophysical Research: Solid Earth, 118(5), 2460- 2473.

228

Watcham, E.P., Shennan, I., Barlow, N.L.M., 2013. Scale considerations in using diatoms as indicators of sea-level change: lessons from Alaska. Quaternary Science Reviews 28(2), 165- 179.

Whiting, M. C., & Mcintire, C. D. (1985). An Investigation Of Distributional Patterns In The Diatom Flora Of Netarts Bay, Oregon, By Correspondence Analysis. Journal Of Phycology, 21(4), 655-661.

Williams, H. F., 2009. Stratigraphy, sedimentology, and microfossil content of Hurricane Rita storm surge deposits in southwest Louisiana. Journal of Coastal Research, 1041-1051.

Williams, H. F., Hutchinson, I., & Nelson, A. R., 2005. Multiple sources for late-Holocene tsunamis at Discovery Bay, Washington State, USA. The Holocene, 15(1), 60-73.

Wilson, R., Hemphill-Haley, E., Jaffe, B., Richmond, B., Peters, R., Graehl, N., Kelsey, H., Leeper, R., Watt, S., McGann, M., Hoirup, D., Chague-Goff, C., Goff, J., Caldwell, D., and Loofbourrow, C., 2014, The search for geologic evidence of distant-source tsunamis using new field data in California, chap. C of Ross, S.L., and Jones, L.M., eds., The SAFRR (Science Application for Risk Reduction) tsunami scenario: U.S. Geological Survey Open-File Report 2013–1170, 214 p., http://dx.doi./of/20131170/c/.

Witter, R. C., Kelsey, H.M., and Hemphill-Haley, E., 2001. Pacific storms, El Nino and tsunamis: competing mechanisms for sand deposition in a coastal marsh, Euchre Creek, Oregon. Journal of Coastal Research, 563-583.

Witter, R. C., Kelsey, H.M., and Hemphill-Haley, E., 2003. Great Cascadia earthquakes and tsunamis of the past 6700 years, Coquille River estuary, southern coastal Oregon. Geological Society of America Bulletin 115, no. 10: 1289-1306.

Woodruff, J.D., Donnelly, J.P., Okusu, A., 2009. Exploring typhoon variability over the mid-to- late Holocene: evidence of extreme coastal flooding from Kamikoshiki, Japan. Quaternary Science Reviews 28, 1774-1785.

Woodroffe, S. A., and Long, A. J., 2010. Reconstructing recent relative sea-level changes in West Greenland: Local diatom-based transfer functions are superior to regional models. Quaternary International, 221(1), 91-103.

Yeats, R.S., Sieh, K.E., Allen, C.R., 1997. The geology of earthquakes. Oxford University Press, Oxford, 568 pp.

Yoshida, T. Jogan Ju’uichi Nen Mutsu Fujou no Shindou Kouitsu [Quake and flood in 11th year of Jogan era] Rekishi Chiri 8, 1033-1040 (1906) [in Japanese]

229

Zong, Y.Q., Tooley, M.J., 1996. Holocene sea-level changes and crustal movements in Morecambe Bay, northwest England. Journal of Quaternary Science 11, 43-58.

Zong, Y., 1997. Implications of Paralia sulcata abundance in Scottish isolation basins. Diatom Research, 12(1), 125-150.

Zong, Y., & Horton, B. P. (1998). Diatom zones across intertidal flats and coastal saltmarshes in Britain. Diatom Research, 13(2), 375-394.

Zong, Y.Q., Horton, B.P., 1999. Diatom-based tidal-level transfer functions as an aid in reconstructing Quaternary history of sea-level movements in the UK. Journal of Quaternary Science 14, 153–167.

Zong, Y.Q., Shennan, I., Combellick, R.A., Hamilton, S.L., Rutherford, M.M., 2003. Microfossil evidence for land movements associated with the 1964 AD Alaska earthquake. Holocene 13, 7-20.

230

INDEX

1 DCA biplot, 113, 114, 115, 116, 118, 119, 120, 121, 123, 124, 125, 126, 127, 128, 129 137Cs, 104, 105, 109, 119, 124, 129, 153 deposit, 15, 16, 17, 29, 30, 34, 35, 44, 45, 46, 67, 104,

132, 143, 173, 180, 181, 194, 198, 200 A diatom, vi, vii, xi, xiii, 8, 9, 10, 11, 12, 13, 14, 16, 17, aerophilic, 61, 75, 107, 172 18, 19, 20, 23, 25, 26, 27, 28, 29, 30, 32, 33, 34, 36,

42, 43, 44, 45, 46, 49, 50, 59, 60, 61, 62, 64, 65, 67, B 68, 69, 70, 71, 72, 74, 75, 76, 80, 82, 92, 94, 95, 99,

101, 106, 107, 108, 112, 113, 115, 116, 117, 119, brackish, 7, 15, 18, 19, 20, 27, 29, 31, 35, 49, 58, 61, 120, 121, 122, 123, 124, 125, 126, 127, 129, 130, 62, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 75, 76, 77, 132, 133, 134, 136, 137, 138, 139, 140, 141, 143, 81, 94, 95, 104, 105, 107, 112, 113, 114, 115, 116, 144, 145, 146, 147, 148, 157, 159, 161, 166, 169, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 171, 172, 173, 174, 175, 176, 178, 179, 180, 181, 128, 129, 130, 132, 134, 135, 136, 140, 142, 146, 183, 188, 189, 190, 203, 204, 206, 207 157, 159, 161, 172, 173, 174, 193

C E

earthquake, vi, viii, x, xiii, 1, 2, 3, 5, 6, 7, 9, 10, 11, 12, cluster analysis, xiv, 108, 112, 114, 115, 116, 117, 118, 14, 15, 16, 18, 19, 21, 22, 23, 24, 25, 26, 28, 29, 32, 119, 120, 121, 123, 124, 125, 126, 127, 128, 129, 33, 36, 38, 42, 43, 44, 46, 50, 51, 52, 53, 54, 55, 56, 134, 154, 157, 158, 160 77, 78, 79, 80, 81, 100, 101, 102, 105, 106, 137,

D 138, 139, 140, 141, 142, 143, 144, 147, 166, 169,

172, 174, 175, 176, 177, 178, 179, 185, 186, 188, D10, 60, 64, 65, 66, 67, 68, 69, 70, 71, 73, 88, 109, 189, 191, 192, 193, 194, 196, 197, 198, 199, 200, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 201, 203, 204, 205, 206, 208 120, 121, 122, 123, 124, 125, 126, 127, 128, 129,

132, 153, 164, 171, 174

231 epipelic, 20, 34, 61, 75, 107, 112, 113, 114, 115, 116, L 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, lithologic, 2, 12, 13, 16, 22, 23, 24, 51, 52, 59, 60, 72, 127, 128, 129, 135, 172 74, 80, 99, 100, 107, 109, 130, 139, 171, 179 epiphytic, 20, 34, 61, 75, 107, 112, 113, 115, 117, 122,

123, 125, 127, 129, 136, 172 M epipsammic, 20, 29, 31, 34, 49, 107, 112, 113, 114,

115, 116, 117, 118, 119, 121, 122, 123, 124, 125, marine, vi, 7, 15, 17, 18, 19, 20, 21, 22, 27, 28, 29, 31,

126, 128, 129, 133, 135, 136 34, 35, 47, 49, 50, 61, 64, 65, 66, 67, 68, 69, 70, 71,

72, 73, 75, 77, 81, 94, 95, 107, 112, 113, 114, 115, F 116, 117, 118, 119, 120, 121, 122, 123, 124, 125,

126, 127, 128, 129, 130, 132, 134, 135, 136, 140, freshwater, vi, xii, 7, 15, 18, 19, 20, 26, 27, 28, 29, 31, 142, 145, 146, 157, 159, 161, 172, 173, 174, 175, 35, 47, 49, 51, 53, 56, 58, 61, 62, 65, 66, 68, 69, 70, 176, 187, 190, 191, 193, 200, 202 71, 72, 73, 75, 76, 77, 80, 94, 95, 104, 105, 107, marsh, xii, xiii, xiv, 6, 8, 19, 20, 24, 25, 29, 32, 33, 35, 112, 113, 114, 115, 116, 117, 118, 120, 121, 122, 42, 50, 56, 57, 61, 72, 75, 76, 77, 85, 104, 105, 130, 123, 124, 125, 126, 128, 129, 130, 132, 134, 135, 134, 135, 136, 170, 179, 192, 194, 198, 199, 200, 136, 139, 141, 146, 157, 159, 161, 172, 173, 174, 203, 204, 206, 207 193, 205 Microfossils, 7, 14, 16, 31, 101, 201 G P grain-size, 12, 13, 59, 65, 66, 67, 71, 88, 106, 108, planktonic, vi, 21, 27, 30, 31, 33, 49, 50, 61, 62, 64, 130, 144, 153, 181 65, 66, 67, 69, 71, 72, 73, 94, 95, 107, 112, 113,

H 114, 115, 116, 117, 118, 119, 120, 121, 122, 123,

124, 125, 126, 127, 128, 129, 130, 133, 135, 136, Historical, 49, 51, 58, 137, 162, 197 142, 145, 154, 157, 158, 159, 160, 161, 172, 173,

175, 176

232

Q slip, 2, 4, 5, 6, 10, 22, 46, 51, 60, 102, 103, 107, 137, 138, 141, 142, 144, 169, 197, 198 Quidico, vii, x, xi, 99, 102, 103, 105, 106, 107, 109, stratigraphic, vi, 2, 6, 11, 12, 13, 14, 22, 36, 50, 54, 59, 110, 111, 125, 130, 131, 132, 133, 134, 136, 137, 61, 72, 78, 88, 90, 91, 96, 101, 104, 105, 106, 109, 138, 139, 140, 141, 142, 144, 145, 146, 147, 150, 130, 131, 137, 141, 143, 144, 146, 147, 148, 151, 151, 170, 171, 194 153, 154, 158, 160, 162, 165, 170, 174, 175, 176, Quintero, vi, ix, 54, 56, 57, 58, 72, 76, 77, 78, 79, 80, 185, 194, 199 85, 87, 90, 170, 171, 176, 178 T R Tirúa, vii, x, xi, 99, 102, 103, 104, 105, 106, 107, 109, radiocarbon, xii, xv, 12, 13, 28, 50, 59, 62, 71, 80, 87, 110, 111, 119, 125, 130, 131, 133, 134, 136, 137, 88, 96, 102, 105, 106, 109, 128, 130, 142, 153, 165, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 166, 178, 188, 189, 201 150, 151, 166, 170, 171, 181 recurrence, vi, xi, 3, 5, 6, 10, 11, 12, 15, 26, 28, 51, 54, tsunami, vi, vii, viii, ix, x, xi, 1, 3, 5, 6, 7, 9, 10, 11, 12, 63, 79, 93, 110, 144, 145, 147, 169, 175, 176, 198, 14, 15, 16, 17, 18, 19, 20, 22, 26, 27, 28, 29, 30, 34, 199, 201, 203 35, 36, 38, 44, 45, 46, 50, 52, 53, 54, 55, 58, 63, 72,

73, 74, 77, 78, 80, 84, 93, 94, 95, 99, 100, 101, 104, S 105, 106, 110, 119, 125, 130, 131, 132, 133, 136, sand bed, 14, 28, 30, 35, 65, 66, 67, 69, 70, 71, 73, 74, 137, 139, 140, 141, 142, 143, 144, 145, 146, 147,

80, 96, 105, 106, 133, 137, 140, 158, 160, 162, 165, 157, 159, 161, 166, 169, 170, 171, 172, 173, 174,

173, 174, 180 175, 176, 177, 178, 180, 181, 182, 185, 187, 188, sea level, x, 2, 16, 32, 50, 52, 57, 77, 79, 99, 100, 177, 190, 191, 192, 193, 194, 195, 196, 198, 199, 200,

199, 200, 203, 204 201, 203, 204, 205, 206, 207

Tychoplanktonic, 61, 98, 108, 164, 184

233