MACROALGAE AND THEIR ASSOCIATED COMMUNITIES: 
CHEMICAL, CONSERVATION, AND TROPHIC ECOLOGY
by
ROSS DOUGLAS BYRON WHIPPO
A DISSERTATION
Presented to the Department of Biology
and the Division of Graduate Studies of the University of 
Oregon in partial fulfillment of the requirements
for the degree of
Doctor of
Philosophy
June 2023
DISSERTATION APPROVAL PAGE
Student: Ross Douglas Byron Whippo
Title: Macroalgae and their Associated Communities: Chemical, Conservation, and Trophic 
Ecology
This dissertation has been accepted and approved in partial fulfillment of the requirements for the 
Doctor of Philosophy degree in the Department of Biology by:
Jeffrey Diez Chair
Aaron W.E. Galloway Advisor
Kelly Sutherland Core Member
Mary I. O’Connor Core Member
Matthew Polizzotto Institutional Representative 
and
Krista Chronister Vice Provost for Graduate Studies
Original approval signatures are on file with the University of Oregon Division of Graduate 
Studies.
Degree awarded June 2023.
ii
© 2023 Ross Douglas Byron Whippo
This work is licensed under a Creative Commons BY Attribution License
Chapter III originally appeared on the ODFW Marine Reserves Website
Open access under Creative Commons BY Attribution as Whippo et al. 2022
iii
DISSERTATION ABSTRACT
Ross Douglas Byron Whippo 
Doctor of Philosophy 
Department of Biology            
June 2023
Title: Macroalgae and their Associated Communities: Chemical, Conservation, and Trophic 
Ecology
 
Marine macroalgae are ubiquitous across coastal oceans worldwide and provide critical 
habitat and services for diverse assemblages of organisms as well as services including 
fisheries production, carbon sequestration, and nutrient cycling. The effects of climate change
are altering multiple aspects of macroalgal community ecology including food web structure, 
organismal diversity, and community resilience. In this dissertation, I approach macroalgal 
community ecology on multiple scales, across broad geographic space, and through various 
analytical lenses to capture key insights into the functioning of macroalgal forests. In Chapter 
II I identify the unique fatty acid profiles and stable isotope content of 31 Antarctic 
macroalgae. While the phylogenetic differentiation driven by fatty acids has a stronger 
influence on distinguishing Antarctic macroalgae, the added dimension of stable isotopes can 
likely make the combination of the two approaches particularly powerful in the application of 
food web studies. In Chapter III I provide a summary of the effects of conservation areas on 
algal assemblages. I find that all targeted kelp species are observed across the marine reserves
iv
but their presence varies among sites and years. In Chapter IV I describe the feeding 
preferences of the sunflower seastar through meta-analysis and a cafeteria experiment. A total
of 114 prey taxa are reported across all studies, with bivalves and urchins tending to be the 
primary observed prey items. This agrees with our cafeteria experiments that find Pycnopodia
tended to prefer green and purple urchins, and mussels, although the quantity of each prey 
type consumed is highly variable. In Chapter V I provide evidence for non-consumptive 
effects of the sunflower seastar on grazing sea urchins. I find that the presence of a 
waterborne Pycnopodia cue reduces the grazing rate of fed urchins by 50% over short (~24 h)
time scales. In contrast, starved urchins consume kelp and do not exhibit an escape response 
in the presence of a Pycnopodia cue. This chapter highlights a trait-mediated indirect 
interaction between Pycnopodia, S. purpuratus and kelp, and how the urchin response to a 
predator cue may differ based on urchin metabolic conditions. This dissertation includes 
previously published and unpublished co-authored material.
v
CURRICULUM VITAE 
NAME OF AUTHOR: Ross Douglas Byron Whippo
GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED:
University of Oregon, Eugene, Oregon, USA
University of British Columbia, Vancouver, British Columbia, Canada
University of Washington, Seattle, Washington, USA
Seattle Central Community College, Seattle, Washington, USA
DEGREES AWARDED:
Doctor of Philosophy, Biology, 2023, University of Oregon 
Master of Science, Zoology, 2013, University of British Columbia
Bachelor of Science, cum laude, Aquatic and Fishery Sciences, 2011, University of 
Washington
Associate of Arts, 2009, Seattle Central Community College
AREAS OF SPECIAL INTEREST:
Subtidal Ecology
Coastal Trophic Ecology
PROFESSIONAL EXPERIENCE:
Research Associate, Smithsonian Institution, Washington, DC, 2018-
2021 
Lead Technician, Smithsonian Institution, Edgewater, Maryland, 2015-2018
Dive Technician, Hakai Institute, Calvert Island, British Columbia, 2014
Graduate Researcher, University of British Columbia, Vancouver, British 
Columbia, 2011-2013
Assistant Dive and Safety Officer, Bamfield Marine Science Centre, 
Bamfield, British Columbia, 2013
Dive Technician, Friday Harbor Laboratories, Friday Harbor, Washington, 
vi
2011
Science Interpreter, Seattle Aquarium, Seattle, Washington, 2008-2009
GRANTS, AWARDS, AND HONORS:
National Science Foundation INTERN Award, 2022
PADI Foundation Grant, 2022
Friday Harbor Graduate Research Fellowship, 2019-2022
Oregon Department of Fish and Wildlife Graduate Scholarship, 2019
National Science Foundation - Graduate Research Fellowship, 2018
Raymund Fellowship, UO, 2018-2019
Kevin Flanagan Travel Award, American Academy of Underwater Sciences, 2013
Graduate Research Grant, Bamfield Marine Sciences Centre, 2013
Faculty of Science Graduate Award, UBC, 2011
Dean's Medal, College of the Environment, University of Washington, 2011
 Faculty Merit Award, School of Aquatic & Fishery Sciences, University of 
Washington, 2011
Mary Gates Endowment Scholarship, University of Washington, 2010
PUBLICATIONS:
Galloway AWE, Gravem SA, Kobelt JN, Heady WN, Okamoto DK, Sivitilli 
DM, Saccomanno VR, Hodin J, Whippo R (2023). Sunflower sea star 
predation on urchins can facilitate kelp forest recovery. Proc. R. Soc. B. 
290:20221897. 20221897. http://doi.org/10.1098/rspb.2022.1897
Harper LM, Lefcheck JS, Whippo R, Jones MS, Foltz Z, Duffy JE (2022). 
Blinded by the bright: How species-poor habitats contribute to regional 
biodiversity across a tropical seascape. Diversity and Distribution 28:11. 
doi: 10.1111/ddi.13632
Whippo R, Fields SA, Aylesworth L (2022). SCUBA kelp report. Oregon 
Department of Fish and Wildlife Marine Resources Program. Newport, 
OR. 
https://ecologyreports.oregonmarinereserves.com/Data_Files/7.%20Collab
orations/SCUBA%20Algal%20Swath%20Report/
ODFW_markdown_algae.html
vii
de Pablo LX, Lefcheck JS, Harper L, Paul VJ, Jones S, Whippo R, Seemann J, 
Kline DI, Duffy JE (2021). A doubling of stony coral cover on shallow 
forereefs at Carrie Bow Cay, Belize from 2014 to 2019. Scientific Reports 
11, 19185 doi: 10.1038/s41598-021-96799-2
Ritter CJ, Harper LM, Lefcheck JS, Paul VJ, Whippo R, Jones S, Ogburn MB, 
Duffy JE (2021). Patterns of consumption across a Caribbean seascape: 
Roles of habitat and consumer species composition through time. Frontiers 
in Marine Science 8:742907. doi: 10.3389/fmars.2021.742907
Whalen M, Whippo R, (64 additional authors), and Duffy JE (2020). Climate 
drives the geography of marine consumption by changing predator 
communities. PNAS 117 (45) 28160-28166. doi: 
10.1073/pnas.2005255117.
Whippo R, Knight NS, Prentice C, Cristiani J, Siegle MR, and O'Connor MI 
(2018). Epifaunal diversity patterns within and among seagrass meadows 
suggest landscape‐scale biodiversity processes. Ecosphere 9(11):e02490. 
doi: 10.1002/ecs2.2490
Staaterman E, Ogburn MB, Altieri AH, Brandl SJ, Whippo R, Seemann J, 
Goodison M, Duffy JE (2017). Bioacoustic measurements complement 
visual biodiversity surveys: preliminary evidence from four shallow marine
habitats. Marine Ecology Progress Series 575: 207-215. 
doi:10.3354/meps12188
Lowe AT, Whippo R, Galloway AWE, Britton-Simmons KH, Dethier MN 
(2015). Sedentary urchins influence benthic community composition below
the macroalgal zone. Marine Ecology 36: 129–140. doi: 
10.1111/maec.12124
Whippo R (2013). Seagrass epifaunal communities of Barkley Sound: Epifaunal
diversity varies across small spatial and temporal scales (Master's Thesis). 
doi: 10.13140/RG.2.2.10089.70245
LEADERSHIP AND SERVICE:
Appointed Director (Scholarships), Board of Directors, American Academy of 
Underwater Sciences, 2023-Present, 2018-2019
Land Acknowledgment Committee, University of Washington, Friday Harbor Labs, 2022
Secretary, Board of Directors, American Academy of Underwater Sciences Foundation, 
viii
2021-Present
Director-at-Large (Scholarships, Standards, Diversity), Board of Directors, American 
Academy of Underwater Sciences, 2019-2021
Advisory Committee, Black Girls Dive Foundation, 2018-Present
Committee Member, American Academy of Underwater Sciences, Membership, 
Scholarship, and Standards Committees 2012-2018
Science Volunteer, Friday Harbor Elementary, Expert Day, 2011-2012
Dive Presenter, San Juan Island Schools, Diver for a Day, 2011, 2022
Education Volunteer, Friday Harbor Elementary, Science Outreach Program, 2009
Volunteer, Seattle Aquarium, Life Sciences and Environmental Interpretation, 2008-2010
ix
ACKNOWLEDGMENTS
My PhD research was conducted on the traditional homelands of the original peoples of 
the Pacific Northwest who have lived here since time immemorial. The University of Oregon is 
located on Kalapuya Ilihi, the traditional indigenous homeland of the Kalapuya people whose 
descendants are now citizens of the Confederated Tribes of Grand Ronde Community of Oregon 
and the Confederated Tribes of Siletz Indians of Oregon. The Oregon Institute of Marine 
Biology is located on the traditional lands and waterways of the Confederated Tribes of the 
Coos, Lower Umpqua, and Siuslaw peoples. The field work for Chapter III of my dissertation 
was conducted within the Oregon Marine Reserve system on the ancestral lands and waterways 
of the Confederated Tribes of Grand Ronde, Confederated Tribes of Siletz Indians, Clatsop-
Nehalem Confederated Tribes, Salmon River, Yakina, Tututni, and Tolowa Dee-ni’ peoples. 
Field work for chapters IV-V was conducted at the Friday Harbor Laboratories on the ancestral 
lands and waterways of the Coast Salish, which touches the shared waters of all tribes and bands 
of the central Salish Sea including the Lhaq’temish (Lummi), Lekwungen (Songhees), 
Swinomish, Semiahmoo, Samish, T’sou-ke, WSÁNEĆ, and Jamestown S’Klallam. I recognize 
that my work in these places has only been possible because of the external and internal 
colonization of this region, forced removals, and efforts to render tribal and First Nations 
invisible, assimilated, and/or disappeared. I also recognize the continuing systemic harms from 
settler colonialism, nationalism, and White supremacy, that often  play a role in educational and 
research endeavors. I grew up along the shores of the Salish Sea next to the original stewards of 
the land. For a long time I did not understand what it meant for me to be living and working on 
here. Now I understand, and I will use my science for collaboration, not extraction, for listening, 
x
not telling, for healing the land, not harming it.
*  *  *
 It has been a long and winding path to the completion of this degree, complicated by the 
global COVID pandemic, and social and political strife, and I owe a debt of gratitude to the 
many, many people who helped me get here through it all. To my advisor, Aaron Galloway, 
whom I have known for a long time, thank you for taking a chance on me and bringing me into 
your lab. I appreciate the opportunities you have provided, and the freedom to pursue my 
interests. To my other committee members: Jeff Diez, Kelly Sutherland, Matt Polizzotto, and 
Mary O’Connor, thank you for your support and your guidance. I also would like to recognize 
Krista McGuire, my former committee chair, for her insights early in my journey as a PhD 
student. I also owe a special debt of gratitude to Katrin Iken, whom I consider an ‘unofficial’ 
committee member, for her ceaseless support, feedback, encouragement, and wine and cheese 
working lunches. I would not have made it here without her. 
To all the folks at OIMB working to keep things up and running over the years, I thank 
you. Particularly Laura Screen, Tammy Trost, James Johnson, and Knute Nemeth. I’m also 
grateful for the patient assistance of Trish Mace who has always been there to lend a hand, even 
when I’m running behind. I am grateful for diving assistance from our DSO Kevin Buch and 
assistant DSO Taylor Eaton. My first year at OIMB was funded by the University of Oregon’s 
Raymund Fellowship, a portion of my second was funded by the Charleston Marine Life Center 
Research Apprenticeship, and subsequent years were funded by the National Science 
Foundation’s Graduate Research Fellowship. All of these fellowships gave me the freedom to 
explore my intellectual interests and were essential to my completion of this program. I am 
xi
grateful to the funders for their support. 
My field work has spanned hemispheres, from the icy waters of Antarctica, to the 
turbulent waters of the Salish Sea, and in each place I benefited from the support and expertise of
those I was lucky enough to collaborate with. In Antarctica I worked with the science dream 
team of Katrin Iken, Chuck Amsler, Alex Lowe, Julie Schram, Andrew Klein, Sabrina Heiser, 
Maggie Amsler, and Aaron Galloway. Thank you for the experience of a lifetime, as well as the 
surprise birthday party in the middle of the Drake Passage. We were supported by Captain Zsolt 
Esztergomi and his invaluable R/V Laurence M. Gould crew, and the Antarctic Support Contract
team of Sean Bercaw, Chuck Holloway, Linnah Neidel, Kevin Pedigo, Gina Pikton, and Rich 
Thompson. 
In Oregon, I was fortunate to have Lindsay Aylesworth and Stephanie Fields at the 
Oregon Department of Fish and Wildlife provide key support just as the pandemic took off, 
which allowed me access to the ODFW Marine Reserves dataset, the synthesis of which is now 
Chapter III in my dissertation. Thanks also to Jenna Clos and Doug Batson for training in the 
survey methods we used (pre-pandemic) to gather the data. This work was supported by the 
Oregon Department of Fish and Wildlife Graduate Fellowship.
Originally, my dissertation was going to be on a completely different topic, but as the 
world shut down it was clear that it would not be possible for me to continue that line of inquiry. 
This is where the Friday Harbor Laboratories at the University of Washington came to the 
rescue. FHL has been my second home for over a decade, and I am eternally grateful to the 
assistance I have received from the labs over the years. In particular, I thank Megan Dethier for 
her generosity, tenacity, and faith in my work. With Bernadette Holthuis, she kept the labs safe 
xii
and open even at the darkest points of COVID, which is the only reason I have Chapters IV and 
V. That research was funded by the Richard R. and Megumi F. Strathmann Endowed Fellowship 
and the Frederic H. Nichols and Kirstin M. Clark Nichols Endowed Graduate Fellowship 
through the University of Washington, and a grant from the PADI Foundation. I also thank the 
entire FHL office and research staff, especially: David Duggins, Pema Kitaeff, Kristy Kull, 
Tommy Pieples, Stephanie Zamora, Mason Wiley, Diana Pieples, Emily Carrington, Morgan 
Johnston, Adam Summers, and Alisa Schoultz. My time at FHL was further enriched by a 
collaboration with the Smithsonian’s MarineGEO Program, the Samish Indian Nation 
Department of Natural Resources, and the Seattle Aquarium. I’m thankful for assistance 
provided by Emmett Duffy, Todd Woodard, Jennie De La Cruz, Charlie Donahue, Zachary 
Randell, and Megan Williams.
Of course so much of what I have done was with the help and support of my fellow 
students and post-docs. At OIMB I am grateful for the wonderful crew of undergrad and 
graduate students who made my time there so much better. I owe special thanks to my lab mates,
past and present: Reyn Yoshioka for his guidance navigating the graduate school and dive-buddy
assistance; Jessie Masterman for her field assistance, endlessly salty humor, and ‘working’ 
vacations; Samantha Persad for the gift of Ora Oshun; Kendall Smith for advice in trying times; 
Rhoda Tetteh for conference shenanigans; Hannah Hayes and Steven Manos because “I don’t 
really wanna do the work today”; Julie Schram for insight into the complicated world of fatty 
acids; Ethan Porter-Hughes for watching endless urchin videos for me; and all the other 
undergrads, graduates, and post-docs that have called CTEL lab home. Also, a big thanks to all 
the other OIMBers including: Nicole Nakata, Caitlin Plowman, Lauren Rice, Christina Ellison, 
xiii
Avery Calhoun, and Erin Jezuit for commiseration, support, and inspiration. Special thanks to 
Joanna Lyle and Dominic Librie for hosting social hours. 
I was also assisted by various students and post-docs at FHL and beyond who lent 
assistance in the lab and in the field. I’m especially grateful to: Tim Dencker, Eliza Heery, Julia 
Kobelt, Marco Corrales, Kindall Murie, Robin Fales, Mo Turner, Abigail Ames, Kayla Hall, 
Karly Cohen, Izzy Graham, Carla Narvaez, Brooke Weigel, Dominic Sivitilli, Willem 
Weertman, Aaron Ninokawa, Jack Litle, Cassandra Donatelli, and so many others. 
Finally, to my friends and family who encouraged me and helped me believe in myself 
even when nothing seemed to go right. I’m especially grateful for my partner Daniel 
Dobrosielski for not only tolerating a boyfriend in a PhD program, but for actually agreeing to 
marry him! Daniel, I couldn’t have done this without you and I’m grateful everyday to have you 
in my life. 
xiv
For Doctor Jones who saw me through to the very end. I’ll miss you, buddy. 
xv
TABLE OF CONTENTS
Chapter Page
I. INTRODUCTION...........................................................................................................1
II. FATTY ACID PROFILES AND STABLE ISOTOPE COMPOSITION OF ANTARCTIC 
MACROALGAE: A BASELINE FOR A COMBINED BIOMARKER APPROACH IN  
FOOD WEB STUDIES...................................................................................................6
1. Introduction.......................................................................................................6
2. Methods.............................................................................................................10
     2.1 Site Selection..............................................................................................10
     2.2 Field Collections.........................................................................................12
     2.3 Fatty Acid Analysis....................................................................................12
          2.3.1 Extraction............................................................................................13
          2.3.2 Transesterification...............................................................................13
          2.3.3 Mass Spectrometry..............................................................................14
     2.4 Stable Isotope Analysis...............................................................................15
     2.5 Statistical Analysis......................................................................................16
3. Results...............................................................................................................17
4. Discussion.........................................................................................................33
III. MULTI-YEAR MONITORING OF MACROALGAL ABUNDANCE AND DIVERSITY 
WITHIN THE OREGON MARINE RESERVE SYSTEM AND 
COMPARISON AREAS.................................................................................................43
1. Introduction.......................................................................................................43
     1.1 Knowledge Gaps.........................................................................................46
     1.2 Research Questions.....................................................................................48
2. Takeaways.........................................................................................................48
     2.1 Kelp Observations......................................................................................48
     2.2 Declines And Methodology........................................................................48
     2.3 Limitations Of SCUBA..............................................................................50
     2.4 Abiotic Parameters......................................................................................51
     2.5 Data As Snapshots......................................................................................51
3. Recommendations.............................................................................................52
     3.1 Permanent Transects...................................................................................52
     3.2 Shallower Transects....................................................................................52
     3.3 Survey Peak Production..............................................................................52
     3.4 UAV Surveys..............................................................................................53
4. Methods.............................................................................................................53
     4.1 Field Surveys..............................................................................................53
     4.2 Data Analysis..............................................................................................55
xvi
5. Results...............................................................................................................56
     5.1 Data Overview............................................................................................56
     5.2 Protocol Comparison..................................................................................56
     5.3 Data By Site................................................................................................57
          5.3.1 Redfish Rocks.....................................................................................57
          5.3.2 Otter Rock...........................................................................................64
          5.3.3 Cascade Head......................................................................................70
          5.3.4 Cape Falcon........................................................................................78
IV. FEEDING PREFERENCES OF THE SUNFLOWER SEA STAR PYCNOPODIA 
HELIANTHOIDES..........................................................................................................85
1. Introduction.......................................................................................................85
2. Methods.............................................................................................................89
     2.1 Meta-Analysis.............................................................................................89
     2.2 Cafeteria Experiment..................................................................................90
     2.3 Statistical Analyses.....................................................................................92
3. Results...............................................................................................................96
     3.1 Meta-Analysis.............................................................................................96
     3.2 Cafeteria Experiment..................................................................................98
4. Discussion.........................................................................................................104
V. THE SUNFLOWER SEA STAR PYCNOPODIA HELIANTHOIDES REDUCES SEA 
URCHIN GRAZING BEHAVIOR THROUGH NON-CONSUMPTIVE EFFECTS..........111
1. Introduction.......................................................................................................111
2. Methods.............................................................................................................115
     2.1 Feeding Experiment Organism Collections And Care................................115
     2.2 Feeding Experiment Setup..........................................................................116
     2.3 Movement Experiment Organism Collections And Care............................118
     2.4 Movement Experiment Setup.....................................................................119
     2.5 Statistical Analysis......................................................................................121
          2.5.1 Feeding Experiments...........................................................................121
          2.5.2 Movement Experiment........................................................................121
3. Results...............................................................................................................122
     3.1 Feeding Experiment....................................................................................122
     3.2 Movement Experiment...............................................................................123
4. Discussion.........................................................................................................127
VI. CONCLUSION...............................................................................................................135
     REFERENCES CITED...................................................................................................139
xvii
LIST OF FIGURES
Figure Page
         2.1 Map Of Western Antarctic Peninsula.......................................................................11
2.2 Fatty Acid Proportions.........................................................................................18
2.3 Fatty Acid Cluster Diagram...................................................................................21
2.4 Fatty Acid nMDS...................................................................................................22
2.5 Fatty Acid PCA......................................................................................................23
2.6 Reduced Fatty Acid PCA........................................................................................26
2.7 Stable Isotopes Biplot.............................................................................................27
2.8 Stable Isotopes Cluster Diagram..............................................................................29
2.9 Stable Isotopes nMDS............................................................................................30
2.10 Stable Isotope PCA...............................................................................................31
2.11 Fatty Acid And And Stable Isotope PCA...............................................................32
2.12 Reduced Fatty Acid And Stable Isotope PCA........................................................32
3.1 Survey Depths........................................................................................................57
3.2 Redfish Rocks Map................................................................................................58
3.3 Humbug Comparison Area Map.............................................................................59
3.4 Redfish Rocks Survey Counts.................................................................................60
3.5 Redfish Rocks Combined Kelp Counts....................................................................62
3.6 Redfish Rocks Kelp Counts....................................................................................64
3.7 Otter Rock Map......................................................................................................64
3.8 Otter Rock Survey Counts......................................................................................66
3.9 Otter Rock Combined Kelp Counts.........................................................................68
3.10 Otter Rock Kelp Counts........................................................................................70
3.11 Cascade Head Map...............................................................................................71
3.12 Schooner Creek Comparison Area Map.................................................................72
3.13 Cavalier Comparison Area Map............................................................................73
3.14 Cascade Head Survey Counts................................................................................74
xviii
3.15 Cascade Head Combined Kelp Counts..................................................................76
3.16 Cascade Head Kelp Counts...................................................................................78
3.17 Cape Falcon Map.................................................................................................79
3.18 Low Fishing Pressure Comparison Area Map........................................................80
3.19 Moderate Fishing Pressure Comparison Area Map.................................................81
3.20 High Fishing Pressure Comparison Area Map........................................................82
3.21 Cape Falcon Survey Counts..................................................................................83
4.1 Pycnopodia Wild-Diet Summary............................................................................97
4.2 Prey Proportional Contributions..............................................................................99
4.3 Mean Prey Consumed............................................................................................99
4.4 Urchin Prey Test Diameters..................................................................................100
4.5 Rodgers’ Index Cumulative Consumption.............................................................101
4.6 Prey Diversity And Evenness................................................................................103
5.1 Experimental Setups.............................................................................................118
5.2 Urchin Hourly Consumption Rates........................................................................123
5.3 Proportional Time For Behaviors..........................................................................125
5.4 Urchin Size Versus Consumption..........................................................................132
xix
LIST OF TABLES
Table Page
2.1 Field Collections Sites In The Western Antarctic Peninsula......................................11
2.2 List Of Fatty Acid Standards...................................................................................15
2.3 Fatty Acid And Stable Isotope Values.....................................................................18
2.4 Species List............................................................................................................22
2.5 PERMANOVA Results..........................................................................................24
2.6 SIMPER Analysis..................................................................................................26
2.7 Fatty Acid Rankings...............................................................................................35
3.1 Total Survey Counts...............................................................................................57
3.2 Redfish Rocks Survey Counts.................................................................................60
3.3 Redfish Rocks Survey Depths.................................................................................61
3.4 Redfish Rocks Focal Species Counts.......................................................................63
3.5 Humbug Comparison Area Species Counts.............................................................63
3.6 Otter Rock Survey Counts......................................................................................66
3.7 Otter Rock Survey Depths......................................................................................67
3.8 Otter Rock Focal Species Counts............................................................................68
3.9 Cape Foulweather Comparison Area Species Counts...............................................69
3.10 Cascade Head Survey Counts................................................................................74
3.11 Cascade Head Survey Depths................................................................................75
3.12 Cascade head Focal Species Counts.......................................................................76
3.13 Cavalier And Schooner Creek Comparison Area Species Counts............................77
3.14 Cape Falcon Survey Counts..................................................................................83
3.15 Cape Falcon Survey Depths..................................................................................84
4.1 Prey Items In Meta-Analysis...................................................................................94
4.2 Publications In Meta-Analysis.................................................................................98
4.3 Mean Rogers’ Index Values..................................................................................102
4.4 All Rogers’ Index Values......................................................................................102
5.1 Feeding Mixed-Effects Model Results...................................................................124
5.2 Movement Mixed-Effects Model Results...............................................................126
xx
CHAPTER I
INTRODUCTION
Marine macroalgae are ubiquitous across coastal oceans worldwide and provide critical 
habitat and services for diverse assemblages of organisms (Dayton 1985, Wiencke & Amsler 2012). 
They are frequently defined as ‘foundation species’ (sensu Hughes 2010) that provide complex 
physical structure utilized by a host of organisms as nursery habitat, shelter, and sources of 
nutrition (Steneck et al. 2002). Typically found in nearshore environments and reaching depths 
of up to 50 m, these macroalgae form dense ‘forests’ that foster year-round and seasonal 
communities that can consist of both surface (canopy), mid-water or benthic (understory) 
components. Macroalgal forests composed of kelp (Laminariales) and other seaweeds (non-
Laminariales) are marked by much higher primary productivity than adjacent habitat (Filbee-
Dexter & Scheibling 2014), and can produce up to 1900 g C m-2 y-1 of which up to 35% can be 
released as dissolved organic matter into the water column (Mann 1973, Krumhansl & 
Scheibling 2012). These forests play a critical role in the delivery of nutrients in the form of 
macroalgal detritus to local consumers and adjacent habitats (Polis et al. 1997, Krumhansl & 
Scheibling 2012, Wiencke & Amsler 2012). They also provide additional services including 
fisheries production, carbon sequestration, and nutrient cycling that are estimated to have a value
of $500 billion USD per year globally (Eger et al. 2023). 
The effects of climate change are altering multiple aspects of macroalgal community 
ecology including food web structure, organismal diversity, and community resilience (Harley et 
1
al. 2012). Removal of key players in coastal food webs by marine heat waves, disease 
exacerbated by climate change, and human use of marine habitat all have the potential to effect 
macroalgal assemblages in myriad ways resulting in changes in: distribution (Assis et al. 2017), 
community connectivity (Peña et al. 2017), biomass (O’Connor et al. 2017), food web 
complexity (Byrnes et al. 2011), detrital production (Krumhansl et al. 2014), carbon flow 
(Salomon et al. 2008), and physiology (Brown et al. 2014).
In this dissertation, I approach macroalgal community ecology on multiple scales, across 
broad geographic space, and through various analytical lenses to capture key insights into the 
functioning of macroalgal forests. The chapters are arranged by ecoregion and move from a 
broad (community-wide) to a narrow (species interactions) focus. Chapter II addresses the 
chemical ecology of Antarctic seaweeds, Chapter III describes the community-wide composition 
of macroalgae inside and outside of marine reserves on the West Coast of North America, and 
Chapters IV and V investigate the trophic interactions between important macroalgal forest-
associated species, also along the West Coast of North America. Chapter VI summarizes my 
findings and suggests next-steps as research into the chemical, conservation, and trophic ecology
of macroalgal forests continues into an uncertain future. 
Chapter II, which contains previously unpublished co-authored material,  concerns the 
chemical ecology of Antarctic seaweeds on the Western Antarctica Peninsula (WAP) and 
describes trophic biomarkers including fatty acids and stable isotopes found in algal tissues. The 
WAP supports extensive macroalgal forests that are home to a wide diversity of organisms 
(Wiencke & Amsler 2012). These macroalgal forests are important to the coastal food web (Iken 
et al. 1997, Dunton 2001, Aumack et al. 2017), however, the contribution of these macroalgal to 
Antarctic food webs on larger scales is not well understood. I used biomarker analysis to 
2
quantify a suite of fatty acids and stable isotope values that can be used as a starting place to 
further investigate trophic pathways in Antarctic food webs (Budge et al. 2006, Boecklen et al. 
2011). It summarizes the fatty acid content for 31 macroalgal species, 14 of which have no 
previously published values, and evaluates contributions of both types of biomarkers as 
taxonomic indicators. This chapter will be published with members of the Antarctic field team 
including: Chuck Amsler, Maggie Amsler, Katrin Iken, Julie Schram, Alex Lowe, Andrew 
Klein, Sabrina Heiser, and Aaron Galloway. All members will contribute to the final submitted 
of the manuscript. 
Chapter III, which contains previously published co-authored material, uses long-term 
survey data from the Oregon Marine Reserve System (OMRS) to identify ecological community 
differences between marine reserves and nearby comparison areas, and provide guidance on best 
practices for reserve monitoring going forward. The OMRS was created in 2010 to address 
conservation concerns including human impacts on fish, invertebrate, and algal populations by 
restricting human access, the effects of which were assessed by long-term monitoring. This 
chapter summarises the findings of the macroalgal component of that monitoring effort between 
the reserves and nearby comparison areas for key species, and provides guidance on SCUBA 
survey methods including the use of permanent transects and shallower surveys, that can provide
more consistent and informative data in the future. It was first published with coauthors Lindsay 
Alysworth and Stephanie Fields in the Oregon Department of Fish and Wildlife’s OMRS report 
in 2022. 
Chapter IV assesses the feeding preferences of the important macroalgal forest generalist 
predator, the sunflower sea star Pycnopodia helianthoides. Pycnopodia populations have been 
reduced drastically since 2012 along the West Coast of North America due to sea star wasting 
3
disease, which has been implicated as a major contributing factor to the proliferation of their 
prey—grazing sea urchins—and the subsequent decline of macroalgal forests (Burt et al. 2018, 
Harvell et al. 2019, Hamilton et al. 2021, Galloway et al. 2023). Although Pycnopodia has been 
observed to consume sea urchins, the strength of their preference for this prey item is unknown. 
Understanding how desirable urchins are as prey to Pycnopodia is critical to understanding their 
role in recent macroalgal forest declines, particularly as macroalgal forests are full of potential 
prey items that Pycnopodia might want to eat. This chapter uses meta-analysis of existing 
observations about Pycnopodia diet in the wild, coupled with a cafeteria feeding experiment to 
determine the relative preferences that Pycnopodia display toward common macroalgal forest-
associated prey. It will be published with coauthors Kindall Murie, who provided field and lab 
support, and Aaron Galloway, who provided conceptual support. Both authors will provide 
editorial contributions to the final submitted manuscript. 
Chapter V, which contains unpublished co-authored material, quantifies the non-
consumptive effect Pycnopodia helianthoides on grazing urchin behavior. The interactions 
between predator and prey can have far reaching consequences for the stability and resilience of 
ecological communities (Paine 1966, Kissling & Schleuning 2015). Although the direct 
consumptive effect of Pycnopodia on sea urchins has been quantified (Galloway et al. 2023), the 
non-consumptive effects (i.e., behavioral responses) of urchins to Pycnopodia is not well 
understood. In this chapter I quantified the non-consumptive effect of Pycnopodia on the purple 
sea urchin Strongylocentrotus purpuratus in two experiments to determine the presence and 
strength of these effects on grazing rates and foraging behavior of the urchins.  This chapter will 
be published with coauthors Sarah Gravem who provided conceptual and statistical support, 
Ethan Porter-Hughes who conducted video analyses, and Aaron Galloway who provided 
4
conceptual support. All authors will provide editorial contributions to the final submitted 
manuscript.  
Taken together, these chapters describe important aspects of macroalgal forest ecology 
from Antarctica to the Western North American coast, and are summarized with main 
conclusions in Chapter VI. 
5
CHAPTER II
FATTY ACID PROFILES AND STABLE ISOTOPE COMPOSITION OF ANTARCTIC
MACROALGAE: A BASELINE FOR A COMBINED BIOMARKER APPROACH IN FOOD WEB
STUDIES
This study contains unpublished coauthored material that was conceived by me as part of a larger 
project led by coauthors Charles Amsler, Margaret Amsler, Katrin Iken, James McClintock, Andrew Klein, 
and Aaron Galloway. Field collections were made by myself and coauthors Charles Amsler, Margaret 
Amsler, Katrin Iken, Alex Lowe, Julie Schram, Sabrina Heiser, Aaron Galloway. I was responsible for all 
fatty acid biomarker extraction and lab work, and stable isotope data was prepared by Katrin Iken. All 
statistical analyses were done by myself, and I also wrote the manuscript with editorial input from all 
authors. 
1. INTRODUCTION
Along the coastal Western Antarctic Peninsula (WAP) dense nearshore macroalgal 
forests provide habitat and carbon to a diverse assemblage of marine organisms (Wiencke et al. 
2014, Valdivia et al. 2015, Oliveira et al. 2020). Nearshore marine habitats of the WAP (0-50 m 
depth) contain the greatest macroalgal biomass found on the continent, with approximately 110 
species of an overall diversity of 151 described macroalgal species in Antarctica (Mystikou et al. 
2014, Pellizzari et al. 2017, Oliveira et al. 2020). These macroalgal species make up the majority
of benthic primary production along the WAP (Wiencke et al. 2014, Barnes 2017). It is estimated
that much of the total macroalgal biomass production is channeled annually into the coastal 
system (Quartino & Boraso de Zaixso 2008) or transported to adjacent, deeper habitats as a 
6
spatial subsidy (Fischer & Wiencke 1992). Multiple studies along the WAP have identified the 
overall importance of macroalgae in supporting the coastal food web (e.g., Iken et al. 1997, 
Dunton 2001, Huang et al. 2006, Aumack et al. 2017, Zenteno et al. 2019, Cardona et al. 2021); 
however, the contributions of individual macroalgal species to the Antarctic food web on larger 
scales are less well understood.
The role that macroalgae play in the WAP coastal food web is complex, and is controlled 
by various factors, including physiological differences in macroalgae marked by the presence of 
chemical defense compounds (Amsler et al. 2005, 2009), and environmental factors such as sea 
ice cover (Quartino et al. 2013, Amsler et al. in revision), temperature (Becker et al. 2010, 
Cordone et al. 2018), and light (Deregibus et al. 2016).  Amphipods and gastropods are some of 
the most abundant mesograzers in the Antarctic subtidal and mostly consume small epiphytes 
growing on the larger, often chemically defended macroalgae (Iken 1999, Aumack et al. 2017). 
These small mesograzers are important prey for other predators, and may serve as a link between
benthic primary production and higher trophic levels (e.g., Dauby et al. 2003, Zamzow et al. 
2011, Casaux & Barrera-Oro 2013). However, some consumers prefer chemically-defended 
species to reduce competition for food resources or capitalize on the chemicals for their own 
defenses (Amsler et al. 2013, Heiser et al. 2022). In addition, chemical defenses of Antarctic 
macroalgae can dissipate once algae senesce or die, increasing the palatability to consumers once
the algal material enters the detrital pathway (Reichardt & Dieckmann 1985, Amsler et al. 2012, 
Schram et al. 2019).
Antarctic macroalgae also differ in other attributes important to consumers, such as 
morphological or biochemical characteristics. For example, thallus toughness mediates 
7
palatability for some consumers (Amsler et al. 2005). Also, while nutritional quality of Antarctic 
macroalgae is typically high (Weykam et al. 1996, Amsler et al. 2005), there are differences in 
biochemical content that increase the nutritional quality of some algal species over others (Peters
et al. 2005). One of the nutritional sources that are especially important to consumers are fatty 
acids (FA). The FAs of both Arctic and Antarctic seaweeds are known to differ among 
taxonomic groups (e.g., Graeve et al. 2002, Aumack et al. 2017, Berneira et al. 2020). This 
aligns with well-established FA patterns in macroalgae elsewhere: the FA composition is 
strongly differentiated by the phylogenetic relationships of the groups, with closely related taxa 
having high similarity in FA profiles (Kelly & Scheibling 2012, Galloway et al. 2012). FAs are 
vital to macroalgal metabolic functions, such as maintaining membrane fluidity (Santos et al. 
2017), in addition to contributing critical nutrients to the metabolic processes of most consumers 
(Dalsgaard et al. 2003, Budge et al. 2006).
Some of the more comprehensive studies on FAs in Antarctic macroalgae established that
they contain significant amounts of polyunsaturated fatty acids (PUFAs) (Graeve et al. 2002, 
Santos et al. 2017, Teixeira et al. 2019, Schram et al. 2019, Berneira et al. 2020), especially long-
chained (C18 and C20) PUFAs, which have been identified as key nutritional components in 
aquatic food webs (Ruess & Müller-Navarra 2019). PUFAs are important for growth, production
and fecundity in marine consumers (e.g., Parrish 2009) and enhance the energy transfer 
efficiency among trophic levels in marine systems (Müller-Navarra et al. 2000, Troch et al. 2012,
Eddy et al. 2021). In Antarctic coastal food webs, PUFAs and other FAs from macroalgae are 
effectively assimilated by consumers (Aumack et al. 2017, Schram et al. 2019), likely 
contributing to their nutrition and performance.
8
In addition to providing important nutrition to consumers, FAs can also be used in food 
web studies to trace the consumption of various macroalgal species into consumers (Kelly & 
Scheibling 2012). This applies especially to essential fatty acids (EFAs), which are organic 
molecules required for biological processes that are only synthesized by primary producers and 
must be assimilated into consumer tissues through consumption (Arts et al. 2001, Galloway & 
Winder 2015). These EFAs, along with other FAs, can be used as biomarkers to trace the flow of
trophic resources though a food web and clarify relationships between specific producers and 
consumers (Dalsgaard et al. 2003, Budge et al. 2006). This is enhanced by the fact that FA 
profiles of macroalgal species seem mostly conserved across large geographical scales, making 
them conservative markers for a source species (Khotimchenko et al. 2002). EFAs, in particular, 
can provide important information on what resources a consumer has been assimilating, 
generating quantitative estimates of the consumers diet (Iverson et al. 2004, Galloway & Winder 
2015, Bromaghin 2017, Guerrero & Rogers 2020).
At the same time, a combination of different biomarkers, such as fatty acids with stable 
isotope (SI) information, can be especially powerful in distinguishing different primary producer 
sources in benthic food webs (Hanson et al. 2010, Kelly & Scheibling 2012, Dethier et al. 2013, 
Aumack et al. 2017).  SI are typically used to provide information about food resources (13C/12C; 
δ13C) and trophic position (15N/14N; δ15N) in food web analyses, where the δ13C values of 
different primary producer sources are preserved with minimal fractionation (~1 ‰) between 
consumer trophic levels (DeNiro & Epstein 1978, Caut et al. 2009), and can, thus, be traced 
through the food web (Boecklen et al. 2011). This applies, however, mostly to differences among
larger groupings of primary producer sources, such as phytoplankton and macroalgal production 
9
or terrestrial organic matter, in coastal systems (France 1995, Peterson 1999, Raven et al. 2002). 
However, the ability of SI to distinguish high-resolution (e.g., species-level) taxonomy is limited.
Using SI to distinguish trophic levels in food webs is particularly valuable as FAs typically 
cannot determine trophic level (Kelly & Scheibling 2012). Data collected during SI analyses also
allow for the calculation of the C:N ratio, which is a useful measure of the nutritional value of 
macroalgae as a food source (e.g., Weykam et al. 1996, Peters et al. 2005), and can provide 
additional information about trophic resource use.
Here, we provide FA profiles of 31 species of Antarctic macroalgae across the three 
major macroalgal divisions (Chlorophyta, Ochrophyta, Rhodophyta), significantly expanding the
published profiles of 18 species (Graeve et al. 2002, Santos et al. 2017, Aumack et al. 2017, 
Teixeira et al. 2019, Schram et al. 2019, Berneira et al. 2020). In addition, we provide the carbon
and nitrogen SI values and C:N ratios for most of the same species. Our expanded biomarker 
baselines, especially for FA profiles, is obtained from field-collected specimens, which may 
differ from profiles of some specimens previously analyzed from long-term lab cultures (Graeve 
et al. 2002). We also assess the ability of the two biomarkers types, individually and combined, 
to distinguish macroalgal species and the value of a multiple biomarker (FA and SI) approach in 
Antarctic coastal food web studies. Specifically, we used the various biomarker data sets to test 
how well they could differentiate some taxonomically close and morphologically similar 
macroalgal species, which may contribute differently to coastal food webs.  
2. METHODS
2.1 Site selection
10
Macroalgae were collected in 2019 at 15 sites spanning the WAP from the Joubin Islands
(64°S) in the north to Millerand Island (68°S) in the south (Figure 2.1, Table 2.1), which were 
accessed using small boats from the Antarctic Research and Support Vessel Laurence M. Gould. 
Collections were made opportunistically as part of a larger project assessing the macroalgal 
distribution and their role in the coastal food web along a gradient of sea ice cover (Amsler et al. 
in revision); collections for the present study were aimed to capture the diversity of macroalgal 
species across all sites.
Figure 2.1: Map of study locations along the Western Antarctic Peninsula. Land masses are in 
gray and ocean is in white with off-shelf portions in blue. Study locations are indicated by letters 
according to Table 2.1.
________________________________
Table 2.1: Field collection sites
11
Sheet1
Site ID Latitude (dd) Longitude (dd) Description
A -64.7720 -64.3699 Joubin Islands
B -64.7793 -64.0441 Southeast corner of Bonaparte Point, Anvers Island
C -64.7932 -64.0072 Limitrophe Island
D -64.9002 -63.8531 Wauwermans Islands
E -65.1043 -64.0471 Pleneau Island (off penguin colony on SE corner)
F -65.2402 -64.2309 Argentine Islands near Vernadsky
G -65.5131 -64.4203 Island north of Lahille Island and south of Lippmann Island
I -66.0251 -65.3533 Minnow Islands (part of Fish Islands South of Prospect Point)
J -66.0894 -65.8386 Southern Saffery Islands
L -67.5488 -67.7714 Small island off a small island to the north of Pinero Island
M -68.1758 -67.2682 Randall Rocks off Millerand Island
Y -65.0995 -63.9874 Booth Island
2.2 Field collections
Algal samples were collected between April 28 and May 18 of 2019 using SCUBA, 
between 5 m and 40 m depth. Algal samples were returned to the lab within 1 h of collection 
where they were sorted and identified. Team phycologists verified the identities of the samples 
from morphological and anatomical features, as well as later confirmation from pressed voucher 
specimens, described in Amsler et al. (in revision). A maximum of seven replicates and a 
minimum of one algal tissue sample per species were taken (>20 mg per sample when possible) 
for FA analysis and frozen at -80 °C in microcentrifuge tubes. In most cases, up to three 
replicates per species were collected for bulk SI analysis. Mostly, these samples were from the 
same thalli as for FA analysis but on occasion, specimens were too small and different thalli 
were sampled for the two biomarker approaches. Algal tissues for SI analysis were dried at 60 °C
until constant weight, at least 24 h.
2.3 Fatty acid analysis
Collections of each macroalgal species were freeze-dried in a lyophilizer for 48 h (or 
Page 1
until completely dry) within 8 months of collection. Homogenized tissue was then processed for 
FA extraction at the Oregon Institute of Marine Biology lab (following Taipale et al. 2016).
12
2.3.1 Extraction Approximately 10-20 mg of dried macroalgal tissue was homogenized from 
each sample with a mortar and pestle, suspended in 2 ml of chloroform in a 10 ml graduated 
centrifuge tube under N2, and held at -20 °C overnight. Tissue samples were sufficiently large 
for each replicate that samples did not need to be pooled. After 24 h extraction, 70 μl C19 
standard (GLC Reference Standard 566 C, Nu-Check-Prep, Elysian, MN), 1 ml methanol, and 
0.75 ml 0.9% NaCl water solution were added to each tube containing the tissues. Between each 
step throughout the extraction and transesterification process (see below), sample tubes were 
topped off with N2 to avoid FA oxidation. Samples were then sonicated in an ice water bath for 
10 min. Each tube was vortexed for 10 sec and centrifuged for 5 min at 3000 rpm at 4 °C. The 
separated chloroform solution was transferred to an 8-ml scintillation vial, and the remaining 
material including algal tissues in the centrifuge tube was topped off with 1 ml chloroform and 
re-run through the vortexing, centrifuging, and chloroform removal process. All chloroform 
fractions from one sample were combined in the same scintillation vial. The chloroform solution 
was then evaporated under N2 to approximately 1.5 ml, and 1 ml of this remaining extract was 
moved to a 10-ml graduated centrifuge tube for FA transesterification. This 1-ml chloroform 
subsample of each extract in the centrifuge tubes was evaporated to dryness under N2 and 1 ml 
toluene added to each sample, and the remaining sample volume was dried and total lipid content
gravimetrically determined (data not reported here).
2.3.2 Transesterification To transesterify each sample, 2 ml of 1% H2SO4 in methanol was 
added to each toluene extract and vortexed for 10 sec. Centrifuge tubes were held at 90 °C for 90
min in a water bath. After the incubation period, 1.5 ml 2% KHCO3 and 2 ml hexane were added
to each tube and vortexed for 10 sec. Tubes were then centrifuged for 2 min at 1500 rpm at 4 °C. 
13
Fatty acid methyl esters (FAME) were transferred from the top phase into a new tube for 
evaporation under N2 in a ~30 °C water bath. An additional 2 ml hexane was added to the 
remaining lower phase in the centrifuge tube, vortexed, and centrifuged again as above. The 
FAME were then combined into each respective sample's tube in the water bath. Samples were 
evaporated to dryness, then 1.5 ml hexane was added to each, and the resulting FAME 
transferred to a GC vial and held at -20 °C.
2.3.3 Mass spectrometry The FA content of each sample was quantified in a gas chromatograph 
equipped with a mass spectrometer (GC-MS, Model QP2020, Shimadzu), with a DB-23 column 
29.6 m long, 0.15 μm thick, and a diameter of 0.25 mm, using helium as a carrier gas. For each 
sample, 1 μl of FAME was run through the following heating protocol: 50 °C for 1 min, 
increased by 20 °C per min to 240 °C, and held for 10 min. Retention time and ion content were 
used to identify specific FA in the sample. A calibration curve using four serial dilutions (15 
ng/ml, 50 ng/ml, 100 ng/ml, 250 ng/ml) of known FAs (GLC Reference Standard 566 C, Nu-
Check-Prep, Elysian, MN; Pearson correlation coefficient >0.99) and the major ions for each 
identified sample peak were used to quantify FA concentrations. Total concentrations of FA in 
each sample were extracted using GCMS Postrun Analysis software (v4.41, Shimadzu 
Corporation, Kyoto, Japan). FA quantified were those included in the GLC Reference Standard 
566 C, Nu-Check-Prep, Elysian, MN (Table 2.2) to ensure confident identification. The PUFA 
18:4ω3, which is a diagnostic biomarker for the brown algae, while not part of the standard mix, 
was also included in our analyses as it can be identified with high confidence in chromatogram 
outputs. FAs are presented here according to the omega nomenclature including the number of 
carbons in the FA, number of double-bonds, and the number of carbons from the methyl end to 
14
the first carbon in the double bond closest to the methyl end.
Table 2.2: List of FA targeted in analysis as part of Nu-check Prep standards
Sheet1
Chain      Standard
c8:0 Methyl Octanoate
c9:0 Methyl Nonanoate
c10:0 Methyl Decanoate
c11:0 Methyl Undecanoate
c11:1 Methyl Undecanoate
c12:0 Methyl Laurate
c13:0 Methyl Tridecanoate
c13:1 Methyl Tridecanoate
c14:0 Methyl Myristate
c14:1ω5 Methyl Myristoleate
c15:0 Methyl Pentadecanoate
c16:0 Methyl Palmitate
c16:1 Methyl Palmitoleate
c17:0 Methyl Heptadecanoate
c18:0 Methyl Stearate
c18:1ω9 Methyl Oleate
c18:1ω7 Methyl Vaccenate
c18:2 Methyl Linoleate
c18:3ω6 Methyl Gamma Linolenate
c19:0 Methyl Nonadecanoate
c18:3ω3 Methyl Linolenate
c20:0 Methyl Arachidate
c20:1 Methyl 11-Eicosenoate
c20:2 Methyl 11-14 Eicosadienoate
c20:3ω6 Methyl Homogamma Linolenate
c20:4 Methyl Arachidonate
c20:3ω3 Methyl Eicosatrienoate
c22:0 Methyl Behenate
c22:1 Methyl Erucate
c20:5 Methyl Eicosapentaenoate
c22:2 Methyl Docosadienoate
c22:3 Methyl Docosatrienoate
c22:4 Methyl Docosatetraenoate
c22:5ω6 Methyl Docosapentaenoate
c22:5ω3 Methyl Docosapentaenoate
c22:6 Methyl Docosahexaenoate
c23:0 Methyl Tricosanoate
c24:0 Methyl Lignocerate
c24:1 Methyl Nervonate
2.4 Stable isotope analysis
Dried macroalgal samples were homogenized to a powder and the carbon and nitrogen SI
composition determined at the Alaska Stable Isotope Facility (ASIF) at the University of Alaska 
Page 1
Fairbanks. Analyses were conducted on a continuous-flow isotope ratio mass spectrometry 
15
(CFIRMS) using a Thermo Scientific Flash 2000 elemental analyzer and Thermo Scientific 
Conflo IV interfaced with a Thermo Scientific DeltaVPlus Mass Spectrometer. Approximately 
0.8-1.2 mg macroalgal material were weighed into tin capsules for analysis. Results are 
expressed as conventional δ notation in parts per thousand (‰) according to the following 
equation: δX (‰) = ([Rsample/Rstandard] −1) · 1000, where X is 13C or 15N of the sample and 
R is the corresponding 13C:12C or 15N:14N ratio. Pee Dee Belemnite and atmospheric N2 
served as standards for carbon and nitrogen, respectively. Instrument error at ASIF was <0.2 ‰ 
for both δ13C and δ15N values. The molar ratio of carbon to nitrogen (C:N ratio) was also 
calculated from the outputs of the SI analysis.
2.5 Statistical analysis
Descriptive patterns of mean FA compositional and SI data for each species were 
visualized in R using the tidyverse and viridis packages (Wickham et al. 2019, Garnier et al. 
2021, R Core Team 2022). Additionally, a SIMPER analysis was used to identify FAs that had 
the strongest effect (up to 80% cumulative) on sample differences, which were then used in 
reduced-biomarker PCA and PERMANOVA analyses using the vegan package (Oksanen et al. 
2018). Both sets of biomarkers were visualized using cluster analysis and non-metric 
multidimensional scaling (nMDS). To determine how FA and SI biomarkers drive differences 
among individual samples and taxonomic groupings, a combination of PCA analysis, cluster 
analysis, and PERMANOVA were run for: all samples with FA data; all samples with SI data; 
16
samples for which both types of data were available; select sample pairs of species that are 
morphologically and taxonomically similar, using the vegan and factoextra packages (Oksanen et
al. 2018, Kassambara & Mundt 2020). Differences in C:N ratios (log transformed to meet 
assumptions of normality) among taxonomic divisions were tested with one-way ANOVA and a 
post-hoc Tukey test.
3. RESULTS
We quantified 40 FAs across all samples, 11 of which constituted at least 5% of the mean
proportion of FA content across species (Figure 2.2, Table 2.3). These 11 ‘common’ FAs 
included saturated and unsaturated FAs, including EFAs such as 20:5ω3 (Eicosapentaenoic 
acid), 18:3ω3 (Alpha-linolenic acid), 22:5ω3 (Docosapentaenoic acid), 18:4ω3 (Stearidonic 
acid), and 20:4ω6 (Arachidonic acid). Among the other most common FA was the saturated FA 
(SAFA) 16:0 (Palmitic acid). Macroalgal samples contained between 20.5-44.4% SAFA and 
43.5-70.6% PUFAs. Lambia antarctica, the only green algal representative in our collection, was
particularly rich in 18:3ω3 compared to other taxa (Figure 2). Most brown algal species 
contained greater proportions of 18:1ω7 (Vaccenic acid), 18:1ω9 (Oleic acid), and 18:2ω6 
(Linoleic acid) compared to red algae, which had overall higher proportions of 20:5ω3.
17
Figure 2.2: Mean proportional composition of individual fatty acids that contributed at least 5% 
of total fatty acid content to each macroalgal species. Species are organized in groups by 
division, with the first column being Chlorophyta, followed by Ochrophyta, and lastly 
Rhodophyta. Fatty acids are listed in the legend according to the omega nomenclature, and 
number or replicates of each species are listed above columns. 
________________________________
Table 2.3: Fatty acid composition (mass %), stable isotope values, and C:N ratios of all algae. 
SAFA: sum of saturated fatty acids; MUFA: sum of monounsaturated fatty acids; PUFA: sum of 
polyunsaturated fatty acids.
18
marker_means
Chlorophyta Ochrophyta Rhodophyta
N (FA/SI/Shared) 4/3/3 1/0/- 3/0/- 4/4/4 3/3/3 7/5/5 7/7/7 3/3/3 2/0/- 2/0/- 1/1/1 3/3/1 2/0/- 1/1/1 2/2/1 2/0/- 7/6/6 6/6/3
8:0 0.009 ± 0.006 0.025 0.002 ± 0.001 0.007 ± 0.002 0.002 ± 0.002 0.003 ± 0.003 0.004 ± 0.002 0.006 ± 0.004 0.001 ± 0.001 0.004 ± 0.000 0.003 0.004 ± 0.002 0.001 ± 0.002 0.014 0.007 ± 0.010 0.013 ± 0.006 0.011 ± 0.009 0.026 ± 0.016
9:0 0.193 ± 0.087 0.619 0.044 ± 0.023 0.121 ± 0.086 0.130 ± 0.092 0.077 ± 0.075 0.048 ± 0.050 0.035 ± 0.030 0.036 ± 0.017 0.318 ± 0.131 0.359 0.024 ± 0.023 0.124 ± 0.136 0.524 0.253 ± 0.346 0.319 ± 0.397 0.302 ± 0.151 0.134 ± 0.099
10:0 0.016 ± 0.007 0.058 0.005 ± 0.001 0.011 ± 0.006 0.005 ± 0.002 0.005 ± 0.002 0.005 ± 0.001 0.005 ± 0.001 0.004 ± 0.002 0.013 ± 0.001 0.008 0.006 ± 0.003 0.012 ± 0.006 0.012 0.019 ± 0.026 0.021 ± 0.011 0.019 ± 0.010 0.022 ± 0.009
11:0 0.004 ± 0.002 0.015 0.003 ± 0.001 0.005 ± 0.003 0.024 ± 0.023 0.009 ± 0.003 0.005 ± 0.002 0.007 ± 0.001 0.002 ± 0.000 0.063 ± 0.015 0.005 0.013 ± 0.003 0.007 ± 0.000 0.012 0.021 ± 0.016 0.012 ± 0.004 0.013 ± 0.005 0.014 ± 0.006
11:1 0.029 ± 0.025 0.107 0.004 ± 0.002 0.018 ± 0.013 0.007 ± 0.006 0.009 ± 0.008 0.003 ± 0.002 0.007 ± 0.003 0.006 ± 0.003 0.021 ± 0.008 0.017 0.023 ± 0.009 0.004 ± 0.006 0.016 0.057 ± 0.081 0.037 ± 0.024 0.033 ± 0.026 0.017 ± 0.013
12:0 0.109 ± 0.027 0.105 0.031 ± 0.012 0.027 ± 0.014 0.022 ± 0.016 0.022 ± 0.007 0.023 ± 0.008 0.018 ± 0.004 0.019 ± 0.006 0.091 ± 0.021 0.074 0.063 ± 0.017 1.164 ± 0.031 1.310 0.150 ± 0.069 0.106 ± 0.058 0.099 ± 0.056 0.823 ± 1.097
13:0 0.005 ± 0.007 0.015 0.004 ± 0.004 0.004 ± 0.004 0.006 ± 0.002 0.010 ± 0.010 0.003 ± 0.003 0.001 ± 0.003 0.000 ± 0.000 0.039 ± 0.003 0.034 0.000 ± 0.000 0.039 ± 0.008 0.114 0.022 ± 0.008 0.029 ± 0.011 0.019 ± 0.006 0.027 ± 0.024
13:1 0.013 ± 0.026 0.019 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.003 ± 0.007 0.001 ± 0.002 0.000 ± 0.000 0.000 ± 0.000 0.041 ± 0.058 0.000 0.015 ± 0.025 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.008 ± 0.022 0.023 ± 0.021
14:0 2.109 ± 0.784 10.306 8.862 ± 0.563 3.081 ± 0.445 3.597 ± 0.461 7.921 ± 0.490 5.245 ± 1.129 4.100 ± 0.522 4.163 ± 0.253 4.286 ± 1.108 4.629 1.882 ± 0.363 1.899 ± 0.238 5.374 3.033 ± 0.467 2.715 ± 0.209 2.068 ± 0.260 1.981 ± 0.312
14:1 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.014 ± 0.024 0.113 ± 0.195 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
15:0 0.107 ± 0.037 0.389 0.279 ± 0.052 0.227 ± 0.047 0.177 ± 0.086 0.150 ± 0.028 0.155 ± 0.032 0.286 ± 0.046 0.294 ± 0.070 0.497 ± 0.031 0.481 0.558 ± 0.038 0.239 ± 0.004 0.570 0.292 ± 0.095 0.699 ± 0.102 0.651 ± 0.189 0.280 ± 0.049
16:0 20.289 ± 1.324 18.043 9.554 ± 0.334 17.639 ± 0.636 12.218 ± 1.754 9.600 ± 2.281 12.737 ± 1.987 14.519 ± 1.363 11.918 ± 1.051 22.524 ± 0.961 25.495 27.055 ± 3.224 22.837 ± 0.483 26.452 25.940 ± 0.113 31.933 ± 1.157 31.267 ± 2.239 28.969 ± 3.449
16:1w7c 6.879 ± 4.065 0.479 0.643 ± 0.149 0.148 ± 0.188 6.931 ± 3.619 1.361 ± 1.216 2.150 ± 2.543 3.036 ± 0.676 5.252 ± 2.347 3.736 ± 0.441 3.844 1.867 ± 0.970 0.287 ± 0.035 2.379 5.830 ± 0.394 3.618 ± 0.510 1.156 ± 0.720 0.980 ± 0.715
17:0 0.045 ± 0.008 0.104 0.126 ± 0.011 0.179 ± 0.021 0.103 ± 0.030 0.057 ± 0.009 0.108 ± 0.018 0.129 ± 0.021 0.217 ± 0.045 0.115 ± 0.025 0.274 0.176 ± 0.008 0.100 ± 0.002 0.148 0.069 ± 0.014 0.131 ± 0.058 0.247 ± 0.058 0.109 ± 0.021
17:1w7c 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
18:0 1.352 ± 0.338 3.598 0.623 ± 0.143 1.432 ± 0.705 1.505 ± 0.575 1.171 ± 0.471 2.024 ± 1.019 1.301 ± 0.156 1.573 ± 0.114 0.839 ± 0.091 3.303 0.980 ± 0.234 0.817 ± 0.211 1.633 2.117 ± 0.843 2.068 ± 1.542 1.800 ± 0.488 1.685 ± 0.450
18:1w9c 1.497 ± 0.332 5.762 26.168 ± 1.212 8.071 ± 0.611 6.654 ± 0.839 10.420 ± 3.835 10.797 ± 2.943 9.375 ± 0.884 3.797 ± 0.093 1.792 ± 0.619 2.411 1.128 ± 0.308 1.449 ± 0.347 4.173 2.365 ± 0.803 4.692 ± 0.569 2.953 ± 0.518 2.998 ± 1.486
18:1w7c 7.260 ± 1.992 0.133 0.000 ± 0.000 0.014 ± 0.028 0.241 ± 0.194 0.023 ± 0.041 0.093 ± 0.104 0.043 ± 0.038 0.613 ± 0.020 0.241 ± 0.051 1.345 0.294 ± 0.117 1.508 ± 0.254 1.460 1.469 ± 0.193 0.502 ± 0.081 1.402 ± 0.439 0.906 ± 0.365
18:2w6c 6.280 ± 1.141 14.998 5.735 ± 0.547 6.340 ± 0.900 2.479 ± 1.943 4.857 ± 1.735 5.942 ± 1.268 3.660 ± 1.364 4.178 ± 0.852 0.165 ± 0.075 0.631 0.392 ± 0.142 0.978 ± 0.257 2.873 1.980 ± 0.381 1.110 ± 0.049 4.228 ± 0.515 1.199 ± 0.494
18:3w6 0.463 ± 0.118 0.204 0.612 ± 0.104 0.470 ± 0.097 0.421 ± 0.179 0.689 ± 0.264 0.622 ± 0.184 0.452 ± 0.033 0.591 ± 0.147 0.000 ± 0.000 0.148 0.196 ± 0.021 0.546 ± 0.233 0.249 0.153 ± 0.017 0.029 ± 0.041 0.391 ± 0.216 0.273 ± 0.157
18:3w3 31.597 ± 9.439 7.830 6.897 ± 0.707 11.391 ± 1.574 8.000 ± 1.470 6.142 ± 1.192 8.006 ± 1.608 7.357 ± 0.415 12.183 ± 0.500 0.048 ± 0.067 0.462 0.085 ± 0.147 0.013 ± 0.018 0.000 0.290 ± 0.410 0.000 ± 0.000 0.035 ± 0.049 0.157 ± 0.154
18:4w3 1.201 ± 0.395 11.087 10.251 ± 1.213 9.088 ± 2.233 11.799 ± 2.963 14.448 ± 4.173 10.414 ± 3.030 10.688 ± 2.049 17.063 ± 2.033 0.014 ± 0.020 0.712 0.128 ± 0.149 0.003 ± 0.004 0.000 0.236 ± 0.199 0.099 ± 0.140 0.007 ± 0.013 0.027 ± 0.043
20:0 0.007 ± 0.015 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.067 ± 0.177 0.574 ± 0.838 0.899 ± 0.851 0.000 ± 0.000 0.000 ± 0.000 0.000 0.006 ± 0.010 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.039 ± 0.025
20:1w9 0.375 ± 0.114 0.000 0.015 ± 0.013 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.001 0.000 ± 0.000 0.413 ± 0.374 0.042 ± 0.059 0.697 0.033 ± 0.001 0.100 ± 0.010 0.000 0.041 ± 0.057 0.000 ± 0.000 0.000 ± 0.000 0.012 ± 0.031
20:2w6 0.116 ± 0.024 0.097 0.129 ± 0.007 0.041 ± 0.049 0.097 ± 0.076 0.018 ± 0.018 0.208 ± 0.043 0.175 ± 0.016 0.303 ± 0.159 0.000 ± 0.000 0.216 0.009 ± 0.016 0.647 ± 0.103 0.185 0.124 ± 0.011 0.024 ± 0.034 0.081 ± 0.021 0.145 ± 0.226
20:3w6 0.051 ± 0.034 0.000 0.809 ± 0.156 0.193 ± 0.086 0.251 ± 0.100 0.622 ± 0.374 0.511 ± 0.209 0.416 ± 0.217 1.082 ± 0.074 0.027 ± 0.039 0.144 0.078 ± 0.001 2.103 ± 0.193 0.738 0.139 ± 0.020 0.103 ± 0.145 0.438 ± 0.097 0.669 ± 0.814
20:4w6 2.793 ± 0.444 9.729 10.769 ± 0.223 24.093 ± 2.439 20.896 ± 0.777 21.458 ± 5.326 21.385 ± 2.670 21.305 ± 4.021 14.358 ± 1.076 46.840 ± 0.607 8.720 37.906 ± 2.640 21.884 ± 4.809 11.364 28.846 ± 0.942 30.005 ± 0.256 14.494 ± 1.835 11.778 ± 6.119
20:3w3 0.196 ± 0.055 0.000 0.059 ± 0.102 0.029 ± 0.057 0.000 ± 0.000 0.021 ± 0.026 0.119 ± 0.089 0.126 ± 0.109 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
20:5w3 16.133 ± 8.676 16.084 18.231 ± 1.107 16.853 ± 1.801 19.005 ± 3.282 20.829 ± 1.712 18.673 ± 3.494 14.376 ± 1.002 20.083 ± 1.362 18.199 ± 1.115 42.645 26.555 ± 2.016 43.160 ± 3.265 40.330 26.183 ± 3.090 21.701 ± 0.873 38.163 ± 2.611 46.466 ± 4.741
22:0 0.000 ± 0.000 0.195 0.137 ± 0.123 0.173 ± 0.203 0.018 ± 0.016 0.002 ± 0.005 0.019 ± 0.015 0.010 ± 0.018 0.038 ± 0.001 0.026 ± 0.036 0.000 0.004 ± 0.007 0.044 ± 0.062 0.000 0.000 ± 0.000 0.000 ± 0.000 0.001 ± 0.004 0.011 ± 0.027
22:1w9c 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.029 ± 0.050 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.035 ± 0.087
22:2w6 0.000 ± 0.000 0.000 0.002 ± 0.003 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.014 ± 0.038 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.001
23:0 0.000 ± 0.000 0.000 0.001 ± 0.001 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.006 ± 0.014
22:4w6 0.036 ± 0.072 0.000 0.000 ± 0.000 0.000 ± 0.000 1.094 ± 1.894 0.000 ± 0.000 0.000 ± 0.000 1.394 ± 2.415 0.000 ± 0.000 0.019 ± 0.027 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
22:5w6 0.043 ± 0.050 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.041 ± 0.000 0.000 ± 0.000 0.000 0.010 ± 0.018 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
22:3w6 0.012 ± 0.024 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.001
22:5w3 0.218 ± 0.062 0.000 0.007 ± 0.007 0.292 ± 0.079 4.294 ± 3.721 0.001 ± 0.004 0.018 ± 0.032 6.145 ± 1.311 0.191 ± 0.025 0.000 ± 0.000 0.359 0.052 ± 0.031 0.034 ± 0.048 0.000 0.103 ± 0.146 0.000 ± 0.000 0.020 ± 0.019 0.026 ± 0.064
22:6w3 0.447 ± 0.231 0.000 0.000 ± 0.000 0.000 ± 0.000 0.026 ± 0.044 0.000 ± 0.000 0.053 ± 0.067 0.000 ± 0.000 1.581 ± 0.713 0.000 ± 0.000 2.986 0.297 ± 0.114 0.000 ± 0.000 0.070 0.259 ± 0.090 0.035 ± 0.049 0.092 ± 0.077 0.054 ± 0.132
24:0 0.116 ± 0.135 0.000 0.000 ± 0.000 0.055 ± 0.039 0.000 ± 0.000 0.004 ± 0.006 0.026 ± 0.020 0.016 ± 0.028 0.000 ± 0.000 0.000 ± 0.000 0.000 0.015 ± 0.007 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.023 ± 0.057
24:1w9 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.120 ± 0.104 0.000 ± 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.084 ± 0.206
CN ratio 11.43 ± 6.11 - - 22.15 ± 1.41 12.46 ± 0.33 12.35 ± 0.96 14.41 ± 1.85 18.77 ± 1.51 - - 9.36 6.21 ± 0.51 - 7.46 7.83 ± 0.79 6.93 6.62 ± 0.63 12.59 ± 1.11
d15N 2.16 ± 0.48 - - 1.25 ± 0.61 2.60 ± 2.94 2.82 ± 0.55 4.78 ± 1.60 2.33 ± 0.50 - - 2.33 3.52 ± 0.41 - 2.20 2.22 ± 0.47 2.29 1.57 ± 1.26 4.39 ± 1.02
d13C -31.02 ± 1.37 - - -28.28 ± 1.06 -31.54 ± 1.90 -25.72 ± 1.27 -27.62 ± 1.96 -22.85 ± 0.10 - - -33.23 -22.66 ± 2.85 - -34.89 -36.08 ± 0.67 -35.32 -33.56 ± 0.96 -20.00 ± 4.49
ΣSAFA 24.361 33.472 19.671 22.961 17.807 19.098 20.976 21.332 18.265 28.815 34.665 30.786 27.283 36.163 31.923 38.046 36.497 34.149
ΣMUFA 16.053 6.500 26.830 8.251 13.833 11.816 13.058 12.574 10.081 5.873 8.314 3.509 3.348 8.028 9.762 8.849 5.552 5.055
ΣPUFA 59.586 60.029 53.501 68.790 68.362 69.085 65.965 66.094 71.654 65.312 57.023 65.708 69.368 55.809 58.313 53.106 57.949 60.794
Σn3 49.792 35.001 35.445 37.653 43.124 41.441 37.283 38.692 51.101 18.261 47.164 27.117 43.210 40.400 27.071 21.835 38.317 46.730
Σn6 9.794 25.028 18.056 31.137 25.238 27.644 28.682 27.402 20.553 47.051 9.859 38.591 26.158 15.409 31.242 31.271 19.632 14.064
Σn3/Σn6 5.084 1.398 1.963 1.209 1.709 1.499 1.300 1.412 2.486 0.388 4.784 0.703 1.652 2.622 0.866 0.698 1.952 3.323
ΣSAFA/ΣPUFA 0.409 0.558 0.368 0.334 0.260 0.276 0.318 0.323 0.255 0.441 0.608 0.469 0.393 0.648 0.547 0.716 0.630 0.562
Page 1
Iridaea cordata
Hymenocladiopsis sp.
Gymnogongrus antarcticus
Georgiella confluens
Delisea pulchra
Curdiea racovitzae
Callophyllis atrosanguinea
Ballia callitricha
Austropugetia crassa
Microzonia australe
Himantothallus grandifolius
Desmarestia menziesii
Desmarestia antarctica
Desmarestia anceps
Cystosphaera jacquinotii
Ascoseira mirabilis
Adenocystis utricularis
Lambia antarctica
Table 2.3 (cont.)
marker_means
Rhodophyta
3/4/1 7/6/6 4/3/2 2/3/2 3/3/3 1/3/1 1/0/- 5/5/5 2/2/2 3/3/3 3/3/3 6/7/4 5/5/5
0.004 ± 0.004 0.014 ± 0.004 0.011 ± 0.008 0.003 ± 0.004 0.007 ± 0.005 0.011 0.011 0.008 ± 0.008 0.001 ± 0.002 0.006 ± 0.002 0.010 ± 0.007 0.022 ± 0.015 0.010 ± 0.005
0.323 ± 0.331 0.198 ± 0.062 0.252 ± 0.205 0.915 ± 0.661 0.091 ± 0.022 0.233 0.214 0.050 ± 0.039 0.066 ± 0.091 0.042 ± 0.024 0.358 ± 0.490 0.640 ± 0.554 0.258 ± 0.214
0.008 ± 0.004 0.021 ± 0.017 0.020 ± 0.014 0.022 ± 0.015 0.014 ± 0.010 0.067 0.018 0.007 ± 0.004 0.008 ± 0.002 0.015 ± 0.003 0.018 ± 0.015 0.038 ± 0.030 0.018 ± 0.016
0.005 ± 0.001 0.015 ± 0.009 0.060 ± 0.034 0.014 ± 0.007 0.008 ± 0.005 0.031 0.028 0.005 ± 0.002 0.008 ± 0.005 0.007 ± 0.001 0.007 ± 0.004 0.016 ± 0.008 0.020 ± 0.007
0.017 ± 0.006 0.037 ± 0.019 0.035 ± 0.015 0.013 ± 0.019 0.022 ± 0.018 0.082 0.012 0.015 ± 0.007 0.006 ± 0.008 0.034 ± 0.052 0.029 ± 0.025 0.094 ± 0.131 0.063 ± 0.067
0.056 ± 0.025 0.064 ± 0.036 0.099 ± 0.069 0.159 ± 0.053 0.075 ± 0.047 1.872 0.210 0.062 ± 0.018 0.104 ± 0.047 0.713 ± 0.053 0.040 ± 0.019 0.169 ± 0.054 0.097 ± 0.056
0.006 ± 0.005 0.004 ± 0.008 0.008 ± 0.010 0.019 ± 0.027 0.008 ± 0.009 0.053 0.026 0.005 ± 0.012 0.015 ± 0.007 0.026 ± 0.012 0.000 ± 0.000 0.017 ± 0.014 0.016 ± 0.018
0.000 ± 0.000 0.000 ± 0.000 0.015 ± 0.031 0.000 ± 0.000 0.023 ± 0.040 0.140 0.000 0.005 ± 0.011 0.000 ± 0.000 0.006 ± 0.010 0.000 ± 0.000 0.000 ± 0.000 0.011 ± 0.025
0.871 ± 0.209 2.119 ± 0.552 1.926 ± 0.603 1.104 ± 0.092 7.480 ± 0.720 1.205 2.889 1.771 ± 0.190 1.989 ± 0.235 4.562 ± 0.230 0.203 ± 0.078 2.091 ± 0.833 2.374 ± 0.848
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.161 ± 0.279 0.000 0.000 0.019 ± 0.042 0.000 ± 0.000 0.000 ± 0.000 0.019 ± 0.038 0.000 ± 0.000 0.000 ± 0.000
0.863 ± 0.117 0.314 ± 0.078 0.272 ± 0.093 0.357 ± 0.053 0.577 ± 0.144 0.324 0.500 0.400 ± 0.048 0.440 ± 0.018 0.282 ± 0.073 0.077 ± 0.022 0.456 ± 0.050 0.391 ± 0.120
39.511 ± 1.414 27.836 ± 4.852 27.505 ± 5.379 29.292 ± 0.991 15.067 ± 1.191 28.247 31.036 25.158 ± 1.881 30.954 ± 0.662 22.695 ± 0.141 21.849 ± 0.481 32.667 ± 4.014 30.630 ± 3.008
3.475 ± 0.770 1.314 ± 0.619 0.908 ± 0.597 2.518 ± 0.021 3.943 ± 2.525 7.353 9.415 3.409 ± 5.062 4.648 ± 0.103 3.996 ± 4.935 0.190 ± 0.170 1.751 ± 0.610 0.764 ± 0.398
0.199 ± 0.033 0.127 ± 0.028 0.097 ± 0.034 0.199 ± 0.032 0.074 ± 0.027 0.125 0.154 0.190 ± 0.061 0.100 ± 0.011 0.102 ± 0.021 0.038 ± 0.013 0.192 ± 0.029 0.159 ± 0.039
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.018 ± 0.039
1.995 ± 1.495 2.377 ± 1.348 1.458 ± 0.382 2.811 ± 1.461 1.396 ± 0.838 2.648 1.720 0.896 ± 0.126 1.638 ± 0.195 1.424 ± 0.649 1.539 ± 0.777 3.477 ± 1.935 1.728 ± 1.307
5.548 ± 0.321 1.917 ± 0.769 1.312 ± 0.255 4.366 ± 0.715 3.242 ± 0.593 5.339 4.378 2.024 ± 0.396 5.079 ± 0.563 0.654 ± 0.263 1.201 ± 0.382 4.686 ± 1.543 1.367 ± 0.357
2.809 ± 0.279 11.100 ± 1.628 7.274 ± 2.405 0.302 ± 0.062 1.959 ± 0.389 5.632 2.113 0.509 ± 0.496 11.107 ± 0.494 0.391 ± 0.077 0.928 ± 0.165 0.443 ± 0.328 0.157 ± 0.091
0.472 ± 0.027 2.218 ± 0.229 1.562 ± 0.406 0.507 ± 0.096 0.231 ± 0.127 1.575 0.333 1.602 ± 0.213 0.788 ± 0.011 0.647 ± 0.340 2.945 ± 0.655 0.352 ± 0.050 0.374 ± 0.076
1.670 ± 0.203 0.659 ± 0.045 0.431 ± 0.099 0.000 ± 0.000 0.174 ± 0.066 0.066 0.277 0.132 ± 0.062 0.296 ± 0.034 0.345 ± 0.130 0.185 ± 0.065 0.086 ± 0.079 0.132 ± 0.036
0.112 ± 0.179 0.143 ± 0.110 0.305 ± 0.395 0.000 ± 0.000 0.294 ± 0.207 0.123 0.162 0.084 ± 0.089 0.163 ± 0.109 0.418 ± 0.276 0.147 ± 0.117 0.000 ± 0.000 0.015 ± 0.032
0.509 ± 0.516 0.061 ± 0.082 0.657 ± 0.908 0.000 ± 0.000 0.657 ± 0.836 0.104 0.146 0.165 ± 0.302 0.199 ± 0.089 0.892 ± 0.644 0.084 ± 0.084 0.000 ± 0.000 0.026 ± 0.049
0.000 ± 0.000 0.004 ± 0.011 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.004 ± 0.008 0.000 ± 0.000 0.250 ± 0.047 0.000 ± 0.000 0.000 ± 0.000 0.003 ± 0.006
0.041 ± 0.071 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.336 ± 0.123 0.000 0.000 0.007 ± 0.016 0.144 ± 0.038 0.022 ± 0.019 2.868 ± 0.729 0.000 ± 0.000 0.000 ± 0.000
0.019 ± 0.034 0.207 ± 0.033 0.222 ± 0.072 0.000 ± 0.000 0.000 ± 0.000 0.194 0.000 0.034 ± 0.020 0.041 ± 0.058 0.145 ± 0.016 3.649 ± 0.496 0.000 ± 0.000 0.004 ± 0.009
0.542 ± 0.188 0.254 ± 0.031 0.395 ± 0.038 0.014 ± 0.020 0.048 ± 0.056 0.075 0.050 0.091 ± 0.032 0.521 ± 0.102 4.953 ± 0.728 1.507 ± 0.266 0.000 ± 0.000 0.052 ± 0.048
11.711 ± 0.094 4.635 ± 0.905 17.866 ± 1.444 13.460 ± 1.265 0.649 ± 0.261 12.273 37.177 39.810 ± 19.079 9.069 ± 0.309 13.061 ± 0.572 5.362 ± 1.108 24.865 ± 9.380 47.301 ± 5.332
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.166 ± 0.194 0.000 ± 0.000 0.000 ± 0.000
29.164 ± 1.521 43.312 ± 9.066 37.149 ± 10.768 43.914 ± 1.209 62.783 ± 6.249 31.707 8.481 23.093 ± 13.235 32.212 ± 0.345 43.436 ± 4.584 56.361 ± 4.398 27.937 ± 10.537 13.959 ± 4.987
0.000 ± 0.000 0.044 ± 0.042 0.033 ± 0.024 0.009 ± 0.013 0.049 ± 0.047 0.000 0.000 0.002 ± 0.005 0.011 ± 0.015 0.455 ± 0.296 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.010 ± 0.022 0.000 ± 0.000 0.020 ± 0.034 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.454 ± 0.786 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.136 ± 0.015 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.017 ± 0.024 0.000 ± 0.000 0.001 ± 0.002 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.093 0.000 ± 0.000 0.000 ± 0.000 0.002 ± 0.004 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.000 ± 0.000 0.000 ± 0.000 0.001 ± 0.001 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000
0.016 ± 0.029 0.009 ± 0.024 0.046 ± 0.054 0.000 ± 0.000 0.078 ± 0.075 0.165 0.171 0.103 ± 0.231 0.142 ± 0.134 0.026 ± 0.046 0.085 ± 0.061 0.000 ± 0.000 0.002 ± 0.004
0.054 ± 0.064 0.090 ± 0.194 0.059 ± 0.119 0.000 ± 0.000 0.061 ± 0.106 0.356 0.385 0.218 ± 0.374 0.254 ± 0.041 0.369 ± 0.279 0.000 ± 0.000 0.000 ± 0.000 0.023 ± 0.033
0.000 ± 0.000 0.111 ± 0.082 0.023 ± 0.026 0.000 ± 0.000 0.036 ± 0.062 0.000 0.000 0.008 ± 0.017 0.000 ± 0.000 0.004 ± 0.007 0.000 ± 0.000 0.000 ± 0.000 0.002 ± 0.004
0.000 ± 0.000 0.796 ± 0.996 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 0.000 0.089 ± 0.199 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.000 ± 0.000 0.030 ± 0.068
11.47 ± 8.22 7.61 ± 0.68 5.27 ± 1.02 9.39 ± 1.10 8.81 ± 2.22 7.68 ± 0.85 - 7.07 ± 0.49 9.10 ± 0.58 7.56 ± 1.21 7.49 ± 0.30 14.42 ± 1.17 6.40 ± 1.01
1.39 ± 1.02 3.09 ± 0.48 4.49 ± 0.83 3.16 ± 1.37 3.26 ± 0.34 2.54 ± 1.33 - 2.37 ± 0.68 2.71 ± 1.44 2.96 ± 0.67 6.13 ± 0.37 2.30 ± 0.51 3.72 ± 0.65
-33.46 ± 4.06 -34.35 ± 0.79 -37.28 ± 0.32 -33.98 ± 1.16 -23.70 ± 0.52 -33.72 ± 5.38 - -35.44 ± 2.03 -34.77 ± 0.10 -33.54 ± 1.12 -29.48 ± 0.84 -24.90 ± 1.79 -22.56 ± 3.01
43.841 33.248 31.764 34.904 24.882 34.816 36.806 28.566 35.334 30.583 24.139 39.785 35.706
11.890 15.164 9.544 7.199 9.686 18.546 15.918 6.087 20.984 5.123 5.235 6.974 2.410
44.269 51.588 58.692 57.895 65.429 46.638 47.275 65.349 43.685 64.296 70.627 53.240 61.888
29.855 43.615 38.216 43.914 63.873 32.455 9.345 23.663 32.970 45.141 56.843 27.937 14.025
14.414 7.973 20.476 13.981 1.556 14.183 37.930 41.686 10.715 19.155 13.784 25.303 47.863
2.071 5.470 1.866 3.141 41.049 2.288 0.246 0.568 3.077 2.357 4.124 1.104 0.293
0.990 0.644 0.541 0.603 0.380 0.747 0.779 0.437 0.809 0.476 0.342 0.747 0.577
Page 2
Trematocarpus antarcticus
Sarcopeltis antarctica
Porphyra plocamiestris
Plocamium sp.
Picconiella plumosa
Phyllophora antarctica
Paraglossum salicifolium
Pantoneura plocamioides
Palmaria decipiens
Pachymenia orbicularis
Myriogramme smithii
Myriogramme manginii
Meridionella antarctica
Cluster analysis identified four primary groupings of all algal species based on the full 
suite of FAs quantified (Figure 2.3). These groupings matched division-level differences, with 
the Chlorophyta and Ochrophyta separating out into individual groupings, while the Rhodophyta 
were further subdivided into two primary groupings. 
−1.0
−0.5 oti
i
B qu
in
al cl
H ia
 ja
0.0 y  ca er
a
men lli at ph rcti
ca
ocl ra id c o
s ta
i ho t a
n
p a s  s aCy sti
0.5 is s areDe pli .s es
m sii
ea p D a me
nzie
ulchra ares
ti
sm
1.0 ePa Dchymenia orbicularis
Adenocystis 
utricularis
ta
Iridaea corda Ascoseira mirabilis
osa
 plum
Ge
niella
orgie
co s llaPic
 c
ioid
e G oy nf
m m
luens
oca
no
l ica Sa gp o
ra c
t r ng
neu
c r
nta
r op us
to  a e
 a
lt nn
Pa lla is
ta
 rcti
ion
e an ct ua s
rid rcti
Me ca
Figure 2.3: Cluster diagram of macroalgal species’ average fatty acid profiles based on 40 fatty 
acids analyzed in this study. Colors represent major groupings of algal species by fatty acid 
profiles. Algal species group phylogenetically by Division, with Ochrophyta as one group (pink),
followed by a single species group for Chlorophyta (purple), and two Rhodophyta groups 
(yellow and orange).
The phylogenetic resolution of these division-level cluster groupings was confirmed by nMDS 
and extended to the levels of order and even family (Figure 2.4; see Table 2.4 for taxonomic 
information). Additionally, the full suite of FAs detected supported this division-level separation 
21
Porphyra ploca
P mal im ea sri ta r id secip
Tre iema nto scarpus antarcticus
Austropugetia crassa
Lambia antarctica
Microzonia aust
D raleesma
H resim tiaa  an nct eo pt shallus grandifolius
iumol
ali
cif
 sm
ss
u
uine
a
lo ng
ara
g osa
P  atr
yllis
allo
ph
C tica
hora a
ntarc
Phyllop
Myriogramme smithii
covitza
e
a ra
Curdie
ngin
ii
ma
amm
e .
gr  s
p
io
Myr mi
um
ca
Plo
of the macroalgal species in PCA analysis, with PC1 and PC2 explaining 15.5% and 8.9% of 
variation, respectively (Figure 2.5). 
2D stress = 0.11
A
0.5
0.0 Division
Chlorophyta
Ochrophyta
Rhodophyta
−0.5
−0.4 0.0 0.4
MDS1
B
0.5
Order
Ascoseirales Fucales
Balliales Gigartinales
0.0
Bangiales Gracilariales
Bonnemaisoniales Halymeniales
Bryopsidales Palmariales
−0.5 Ceramiales Plocamiales
Desmarestiales Rhodymeniales
Ectocarpales Syringodermatales
−0.4 0.0 0.4
MDS1
C
Family
0.5
Adenocystaceae Gigartinaceae
Ascoseiraceae Gracilariaceae
Balliaceae Halymeniaceae
0.0 Bangiaceae Kallymeniaceae
Bonnemaisoniaceae Palmariaceae
Bryopsidaceae Phyllophoraceae
Callithamniaceae Plocamiaceae
−0.5 Cystocloniaceae Rhodomelaceae
Delesseriaceae Sarcodiaceae
Desmarestiaceae Seirococcaceae
Fryeellaceae Syringodermataceae
−0.4 0.0 0.4
MDS1
Figure 2.4: nMDS of all macroalgal FA profiles coded by Division (A), Order (B), and Family 
(C).
________________________________
Table 2.4: Macroalgal species collected during the 2019 Western Antarctic Gradients Project 
cruise for FA analysis. Species highlighted in gray do not have a published FA profile.
22
MDS2 MDS2 MDS2
Sheet1
Phylum Order Family Species Reference
Chlorophyta Bryopsidales Bryopsidaceae Lambia antarctica (Graeve et al. 2002)
Ochrophyta Ascoseirales Ascoseiraceae Ascoseira mirabilis (Teixeira et al. 2019)
Desmarestiales Desmarestiaceae Desmarestia anceps (Aumack et al. 2017, Teixeira et al. 2019, Schram et al. 2019)
Desmarestia antarctica (Graeve et al. 2002)
Desmarestia menziesii (Aumack et al. 2017)
Himantothallus grandifolius (Schram et al. 2019)
Ectocarpales Adenocystaceae Adenocystis utricularis (Teixeira et al. 2019, Berneira et al. 2020)
Fucales Seirococcaceae Cystosphaera jacquinotii
Syringodermatales Syringodermataceae Microzonia australe
Rhodophyta Balliales Balliaceae Ballia callitricha
Bangiales Bangiaceae Porphyra plocamiestris
Bonnemaisoniales Bonnemaisoniaceae Delisea pulchra (Schmid et al. 2018)
Ceramiales Callithamniaceae Georgiella confluens (Graeve et al. 2002)
Delesseriaceae Myriogramme smithii (Graeve et al. 2002)
Myriogramme manginii (Berneira et al. 2020)
Pantoneura plocamioides (Graeve et al. 2002)
Paraglossum salicifolium
Rhodomelaceae Picconiella plumosa
Gigartinales Cystocloniaceae Meridionella antarctica
Gigartinaceae Sarcopeltis antarctica (Graeve et al. 2002, Aumack et al. 2017, Berneira et al. 2020)*
Iridaea cordata (Santos et al. 2017)
Kallymeniaceae Austropugetia crassa
Callophyllis atrosanguinea
Phyllophoraceae Gymnogongrus antarcticus
Phyllophora antarctica
Gracilariales Gracilariaceae Curdiea racovitzae (Berneira et al. 2020)
Halymeniales Halymeniaceae Pachymenia orbicularis
Palmariales Palmariaceae Palmaria decipiens (Graeve et al. 2002, Santos et al. 2017, Schram et al. 2019)
Plocamiales Plocamiaceae Plocamium sp. (Santos et al. 2017, Aumack et al. 2017)†
Sarcodiaceae Trematocarpus antarcticus
Rhodymeniales Fryeellaceae Hymenocladiopsis sp. (Graeve et al. 2002)‡
* as Gigartina skottsbergii
† Plocamium cartilagineum
‡ Hymenocladiopsis prolifera (crustigena)
0.4 Species
2200::21ww69 Adenocystis utricularis Lambia antarctica
20:5w3 Ascoseira mirabilis Meridionella antarctica
Austropugetia crassa Microzonia australe
Ballia callitricha Myriogramme manginii
18:1w7c
24:0 20:3w3 Callophyllis atrosanguinea Myriogramme smithii0.2
20:3w6 Curdiea racovitzae Pachymenia orbicularis
24:12w29:2w262:3w6 Cystosphaera jacquinotii Palmaria decipiens
23:0 Delisea pulchra Pantoneura plocamioides
12:0 22:1w9c
2222::05w6 18:3w3 Desmarestia anceps Paraglossum salicifolium
8:0
22:61w43:1 Desmarestia antarctica Phyllophora antarctica
10:0 13:1 16:1w7c 18:3w6 18:2w6c Desmarestia menziesii Picconiella plumosa0.0 16:0
191:0:1 Georgiella confluens Plocamium sp.
18:0 20:0
13:0 17:1w7c 22:4w6 Gymnogongrus antarcticus Porphyra plocamiestris
11:0 22:5w3 Himantothallus grandifolius Sarcopeltis antarctica
Hymenocladiopsis sp. Trematocarpus antarcticus
18:4w3c
14:0 Iridaea cordata
18:1w9c
−0.2
15:0
Division
17:0 Chlorophyta
Ochrophyta
20:4w6 Rhodophyta
−0.2 0.0 0.2
PC1: 15.5%
Figure 2.5: Principle Component Analysis (PCA) of all macroalgal samples based on the full 
suite of 40 fatty acids. Algal species are color-coded in the legend, with phylum indicated by 
shapes. All individual replicates of the algal species are represented. Vectors represent individual
fatty acids, with the length and direction indicatPiangeg 1 the amount of variability and the direction 
23
PC2: 8.9%
associated with each fatty acid according to the principle component axes. Variance explained by
each axis is given as percent along each axis.
Ochrophyta grouped distinctly and tightly along PC1, while Rhodophyta were separated along 
PC1 from the Ochrophyta but had a larger spread along PC2. Species-level differences in full FA
profiles were also detected (PERMANOVA Table 2.5), though there was much overlap among 
species (Figure 2.5). A subset of FA driving differences among samples was determined with 
SIMPER analysis and used in additional PCA analyses to confirm that major division-level 
groupings were still apparent (Figure 2.6, Table 2.5, 2.6). This reduced PCA included primarily 
FAs that were also part of the suite of common FAs that comprised more than 5% of total 
proportional contribution (see Figure 2.2). This reduced PCA explained 43.0% (PC1) and 24.1% 
(PC2) of variation among samples. The pattern observed with the FA subset was similar to that 
seen when the full suite of FA were considered; however, the phylogenetic groupings were even 
more distinct. Overall, division-level differences were supported more strongly than other 
taxonomic levels between both the full FA (PERMANOVA, DF = 2, SS = 4.007, F = 53.363, 
p(perm) = 0.001) and reduced FA (PERMANOVA, DF = 2, SS = 4.065, F = 62.708, p(perm) = 
0.001) analyses, and species-level effects were more clearly pronounced than by order or family 
(Table 2.5).
________________________________
Table 2.5: PERMANOVA table for results across all biomarker and taxonomic analyses. 
24
Sheet1
Biomarkers df SS F P (perm)
FA only 
division
Division 2 4.007 53.363 0.001
Residual 103 3.868
Total 105 7.875
order
Order 15 5.687 15.596 0.001
Residual 90 2.188
Total 105 7.875
family
Family 21 6.568 20.105 0.001
Residual 84 1.307
Total 105 7.875
species
Species 30 7.138 24.221 0.001
Residual 75 0.737
Total 105 7.875
FA reduced 
division
Division 2 4.065 62.708 0.001
Residual 103 3.338
Total 105 7.403
order
Order 15 5.514 17.512 0.001
Residual 90 1.889
Total 105 7.403
family
Family 21 6.336 23.760 0.001
Residual 84 1.067
Total 105 7.403
species
Species 30 6.788 27.610 0.001
Residual 75 0.615
Total 105 7.403
SI only
division
Division 2 0.262 10.794 0.001
Residual 89 1.080
Total 91 1.342
order
Order 11 0.624 6.312 0.001
Residual 80 0.718
Total 91 1.342
family
Family 17 1.016 13.584 0.001
Residual 74 0.326
Total 91 1.342
species
Species 24 1.136 15.410 0.001
Residual 67 0.206
Total 91 1.342
Reduced FA + SI 
division
Division 2 0.298 13.269 0.001
Residual 73 0.818
Total 75 1.116
order
Order 11 0.596 6.667 0.001
Residual 64 0.520
Total 75 1.116
family
Family 17 0.889 13.344 0.001
Residual 58 0.227
Total 75 1.116
species
Species 24 1.002 18.609 0.001
Residual 51 0.114
Total 75 1.116
All FA + SI
division
Division 2 0.301 13.289 0.001
Residual 73 0.826
Total 75 1.126
order
Order 11 0.602 6.677 0.001
Residual 64 0.525
Total 75 1.126
family
Family 17 0.896 13.267 0.001
Residual 58 0.230
Total 75 1.126
species
Species 24 1.005 18.437 0.001
Residual 51 0.116
Total 75 1.121
Page 1
25
Figure 2.6: Principle Component Analysis (PCA) of all macroalgal samples based on a reduced 
set of seven fatty acids as determined by a Similarity Percentage (SIMPER) analysis. See Figure 
2.5 for details on the PCA arrangement.
Table 2.6: SIMPER analysis of contribution of fatty acids to driving differences across all 
samples, through 80% of cumulative variation.
Sheet1
FA average sd cumulative sum
20:5ω3 0.081 0.060 0.226
20:4ω6 0.072 0.059 0.427
16:0 0.048 0.035 0.562
18:3ω3 0.028 0.040 0.639
18:4ω3 0.026 0.029 0.712
18:1ω9 0.023 0.027 0.778
18:1ω7 0.015 0.019 0.820
Stable isotope values were variable across the Rhodophyta and Ochrophyta, with δ13C 
ranging from -37.3‰ to -20.0‰ (mean -30.1‰, Rhodophyta) and -31.5‰ to -22.9‰ (mean -
26
Page 1
27.2‰, Ochrophyta) (Table 2.3).  The δ13C range among the Rhodophyta was particularly large 
and included two broad groupings, one with values highly depleted in 13C and another grouping 
with δ13C values more similar to those of many Ochrophyta (Figure 2.7a,b). The green alga L. 
antarctica and one Ochrophyta (Desmarestia anceps) aligned closely with the 13C-depleted 
Rhodophyta group.  δ15N ranged from 1.4‰ to 6.1‰ (mean 3.0‰, Rhodophyta), and 1.3‰ to 
4.8‰ (mean 3.1‰, Ochrophyta; Figure 2.7a,b, Table 2.3). The single Chlorophyta had a mean 
δ13C of -31.0‰ and  a mean δ15N of 2.2‰. Four macroalgal species were distinct with mean δ15N
values >4‰; these were the Ochrophyta Desmarestia menziesii, and the Rhodophyta Iridaea 
cordata, Myriogramme smithii, and Porphyra plocamiestris. The C:N ratios across all 
macroalgal divisions were variable but overall lower in the Rhodophyta than the other divisions 
with means ranging from 5.3 to 14.4 (mean 8.8, Rhodophyta), 12.3 to 22.2 (mean 15.7, 
Ochrophyta), and 11.4 (Chlorophyta) (Figure 2.7c). Differences in C:N ratio were significantly 
different between Rhodophyta and Ochrophyta, but not between Chlorophyta and the other 
divisions (ANOVA, DF = 2, MS = 392.5, F 30.49, p < 0.0001; Tukey-test, p-adj. < 0.0001). 
________________________________
Figure 2.7: Biplot of mean values for δ15N and δ13C with standard error bars coded by Division 
(A) and Species (B).  Mean C:N ratio (C) is also shown with mean and standard error and each 
species (column) is coded by Division as in panel A. 
27
28
Cluster analysis of stable isotope values (δ13C, δ15N) and the C:N ratio resulted in four groupings 
of algae, although these did not resolve distinctly along the phylogenetic divisions (Figure 2.8, 
see Table 2.4 for taxonomic information). Although taxonomic groupings were not readily 
apparent in nMDS and clusters, PERMANOVA revealed significant Division, Order, and Family
groupings (Figure 2.9, Table 2.6). A PCA analysis of macroalgal samples based only on carbon 
and nitrogen SI values and C:N ratios explained 43.9% (PC1) and 33.7% (PC2) of variation 
among samples (Figure 2.10). This PCA also showed clear division-level groupings although the
separation of Rhodophyta and Ochrophyta along PC1 was not as large as it was in the ordination 
of macroalgal species based on FA analysis (see Figure 2.5). Analyses across all samples 
provided support for all taxonomic groupings; however, species-level effects were strongest 
(PERMANOVA, DF = 24, SS = 1.136, F = 15.410, p(perm) = 0.001; Table 2.5).
−1.0
−0.5 rist
Hi Ir i
es
ma i mn dto a
a
th ea plo
c
a iillu  co a it
h
0.0 s g r mr rda hy e 
s
a
C p my ns tto dsp ih f
a orol P a
m
aera ius yrio
gr
 jacquin M allitric
ha
otii Ballia
 c
0.5 Trematocarpus antarcticus Pachymenia orbicularis
uineang Posa
icc
r os at ns G
niel
phylli ie e
la 
o plumos
Callo ip . D r a
a d
ec  sp e
gi
ri li
e
s lla ium e
a 
lm a
co
 nPa am p
fl
u uc elo lc
n
h sP ra
Figure 2.8: Cluster diagram of macroalgal species based on their average carbon and nitrogen 
stable isotope values as well as their C:N ratio. Colors represent major groupings of algal 
species, with four major groupings identified, which do not group according to phylum.
29
Pantoneura ploca
M myrio ig ora im dm ee smanginii
Hymenocladiopsis sp.
Meridionella antarctica
Lamb
D ia ae nts arm cticaarestia anceps
us
arc
tic
ants 
ong
ru
nogm a
Gy ta
rctic
hora 
an
Phyllo
p
Sarcopeltis antarctica
ntarct
ica
restia
 a sii
Desm
a zieen
tia 
m
are
s
sm
De
Figure 2.9: nMDS of all macroalgal SI profiles coded by Division (A), Order (B), and Family 
(C).
30
Species
d15N
Adenocystis utricularis Lambia antarctica
Ascoseira mirabilis Meridionella antarctica
Austropugetia crassa Microzonia australe
Ballia callitricha Myriogramme manginii
Callophyllis atrosanguinea Myriogramme smithii
0.5 Curdiea racovitzae Pachymenia orbicularis
Cystosphaera jacquinotii Palmaria decipiens
Delisea pulchra Pantoneura plocamioides
Desmarestia anceps Paraglossum salicifolium
Desmarestia antarctica Phyllophora antarctica
Desmarestia menziesii Picconiella plumosa
Georgiella confluens Plocamium sp.
d13C
0.0 Gymnogongrus antarcticus Porphyra plocamiestris
Himantothallus grandifolius Sarcopeltis antarctica
Hymenocladiopsis sp. Trematocarpus antarcticus
Iridaea cordata
Division
Chlorophyta
Ochrophyta
−0.5 CN ratio
Rhodophyta
−0.25 0.00 0.25 0.50 0.75
PC1: 43.9%
Figure 2.10: Principle Component Analysis (PCA) of all macroalgal samples based on the full 
suite of 40 fatty acids as well as the carbon and nitrogen stable isotope values and the C:N ratio. 
See Figure 2.5 for details on the PCA arrangement.
A combined PCA based on either total or reduced FA with carbon and nitrogen SI and 
C:N ratio data from samples that were analyzed for both types of biomarkers (Figure 2.11, 2.12) 
produced very similar patterns as seen in the PCA based solely on total or reduced FAs (Figure 
2.5), with clear delineation of the macroalgal divisions. The combination of both biomarkers 
slightly increased the explanatory power for PC1 (16.5%) and PC2 (9.5%), and created a larger 
separation between Ochrophyta and Rhodophyta (Figure 2.11) compared to the FA-only PCA 
(Figure 2.5). These groupings were maintained, though not as tightly, when comparing the 
reduced FAs combined with SI (Figure 2.12) with the reduced-FA only PCA (Figure 2.6). It 
showed a stronger effect at the species-level (PERMANOVA, DF = 24, SS = 1.002, F = 18.609, 
p(perm) = 0.001) than the division-level, which was in contrast to the reduced FA-only 
groupings that had stronger division than species groupings (Table 2.5).
  
31
PC2: 33.7%
0.4
Species
2200::12w96 Ballia callitricha Meridionella antarctica
Callophyllis atrosanguinea Myriogramme manginii
20:5w3 Cystosphaera jacquinotii Myriogramme smithii
d15N Delisea pulchra Pachymenia orbicularis
0.2 18:1w7c 20:3w3 Desmarestia anceps Palmaria decipiens
202:342w:
2026
2:5w6
:3w6 Desmarestia antarctica Pantoneura plocamioides
24:1w9
22:2w6 Desmarestia menziesii Phyllophora antarctica
Georgiella confluens Picconiella plumosa
222:122w3:6:90wc232:0 18:3w3 Gymnogongrus antarcticus Plocamium sp.
12:0
16:11w47:1c Himantothallus grandifolius Porphyra plocamiestris
0.0 16:0 13:1 18:2w6c8:0 Hymenocladiopsis sp. Sarcopeltis antarctica18:3w6
Iridaea cordata Trematocarpus antarcticus
10:0
11:1131:01:0 17:1w7c 22:4dw163C 20:0 Lambia antarctica
9:0
22:5w3 14:0
18:0 Division
18:4w3c
15:0 ChlorophytaCN ratio
−0.2 Ochrophyta
18:1w9c
Rhodophyta
17:0 20:4w6
−0.2 0.0 0.2
PC1: 16.5%
Figure 2.11: Principle Component Analysis (PCA) of all macroalgal samples based on the full 
suite of 40 fatty acids as well as the carbon and nitrogen stable isotope values and the C:N ratio. 
See Figure 2.5 for details on the PCA arrangement.
Figure 2.12: Principle Component Analysis (PCA) of all macroalgal samples based on the 
reduced suite of 40 fatty acids as well as the carbon and nitrogen stable isotope values and the 
C:N ratio. See Figure 2.5 for details on the PCA arrangement.
32
PC2: 9.5%
The morphologically similar Ochrophyta species Desmarestia menziesii and D. anceps, 
and the Rhodophyta Phyllophora antarctica and Callophyllis atrosanguinea were tested for 
differences based on FA profiles and carbon and nitrogen SI composition. D. menziesii and D. 
anceps were significantly different from one another with regards to both their FA profiles 
(PERMANOVA, DF = 1, SS = 0.03, F = 3.04, p = 0.0425), and SI values (PERMANOVA, DF =
1, SS = 0.01, F = 5.73, p = 0.015). However, P. antarctica and C. atrosanguinea were not 
significantly different from one another with regards to their FA profiles (PERMANOVA, DF = 
1, SS = 0.01, F = 0.40, p =0.7658), but were different in stable isotope values (PERMANOVA, 
DF = 1, SS = 0.07, F = 38.61, p =0.0187).
4. DISCUSSION
In this study, we quantified the FA profiles and SI values of 31 Antarctic algal species to 
assess their biochemical and biogeochemical composition and the possible value of using these 
metrics as biomarkers to differentiate macroalgal species in food web applications. Our work 
extended the number of published Antarctic macroalgal FA profiles by 14 species not covered in 
previous studies (Graeve et al. 2002, Santos et al. 2017, Aumack et al. 2017, Schmid et al. 2018, 
Schram et al. 2019, Berneira et al. 2020). As has been established for macroalgae elsewhere 
(Galloway et al. 2012), the FA profiles of WAP macroalgal species differed phylogenetically by 
the major algal divisions of Chlorophyta, Ochrophyta, and Rhodophyta, and the separation was 
robust also on the order and even the family levels. The grouping by division was preserved 
when only a small subset of FAs was applied, including a number of PUFAs, most of them 
EFAs. SI information was less specific to the taxonomic affiliation of the species, although 
33
divisional-level separation was still noticeable. We suggest that the combination of the two 
biomarker approaches could be especially valuable for differentiating macroalgal species groups 
in food web studies.
Macroalgal species from the WAP varied in their proportional FA compositions, but all 
contained high proportions of PUFAs, often comprising more than 50% of the total FA. The 
Rhodophyta in our study were largely composed of the SAFA 16:0 (mean 27.9%) and the PUFA 
20:5ω3 (mean 35.5%), which is generally comparable to means previously published for 
Antarctic red algae (30.9% and 28.8%, respectively; see Table 2.7; averaged from Graeve et al. 
2002, Santos et al. 2017, Aumack et al. 2017, Schmid et al. 2018, Berneira et al. 2020). 
However, 20:4ω6 was more prominent in our samples (mean 20.0%) compared to previously 
published Antarctic Rhodophyta (9.0%, references as above, Table 2.7), although 20:4ω6 has 
been identified as a major FA in Rhodophyta elsewhere (e.g., Dalsgaard et al. 2003, Kumari et 
al. 2013, Sohrabipour 2019). 16:1ω7 was reported as one of the top five contributing FA for 
Antarctic Rhodophyta in the literature (mean 6.5%, Table 2.7) but was negligible when 
considering the same set of species in our study (Table 2.7). However, when only considering 
new Rhodophyta species not previously investigated for FAs, 16:1ω7 was detectable (mean 
2.7%). Some differences to previous Antarctic macroalgal records could be based on the fact that
FA are reported in percentages instead of absolute concentrations, making these mean values 
dependent on the total number of FAs detected. In our study, we targeted 40 FAs, while other 
studies analyzed between 11 and 35 FAs (Graeve et al. 2002, Santos et al. 2017, Aumack et al. 
2017, Schmid et al. 2018, Berneira et al. 2020), although the identity of the major FA 
contributors should still be similar, with just differences in the exact percentages.
34
Table 2.7: Most abundant FA found in samples ranked by taxonomic division and source of 
values (published = published literature values, literature species = species in our dataset with 
published counterparts, new species = species in our dataset with no published profiles, both = 
overall mean of all species within our dataset). Only FA included in our analyses were extracted 
from the literature. 
Sheet1
Ochrophyta Rhodophyta
this study this study
Rank published* literature species new species both published† literature species new species both
1 20:4ω6 (16.0%) 20:4ω6 (19.5%) 20:4ω6 (20.8%) 20:4ω6 (19.8%) 16:0 (30.9%) 20:5ω3 (40.4%) 20:5ω3 (28.7%) 20:5ω3 (35.5%)
2 20:5ω3 (15.7%) 20:5ω3 (18.6%) 20:5ω3 (17.9%) 20:5ω3 (18.5%) 20:5ω3 (28.8%) 16:0 (27.6%) 16:0 (28.4%) 16:0 (27.9%)
3 16:0 (15.2%) 18:4ω3 (11.8%) 16:0 (15.7%) 16:0 (12.6%) 20:4ω6 (9.0%) 20:4ω6 (14.1%) 20:4ω6 (28.3%) 20:4ω6 (20.0%)
4 18:4ω3 (12.0%) 16:0 (11.8%) 18:4ω3 (11.7%) 18:4ω3 (11.8%) 18:1ω9 (6.6%) 18:1ω7 (3.5%) 18:1ω9 (2.7%) 18:1ω9 (2.7%)
5 18:1ω9 (10.5%) 18:1ω9 (11.7%) 18:3ω3 (11.7%) 18:1ω9 (10.7%) 16:1ω7 (6.5%) 18:1ω9 (2.7%) 16:1ω7 (2.7%) 18:1ω7 (2.6%)
* values from Aumack et al. 2017, Berneira et al. 2020, and Graeve et al. 2002
† values from Aumack et al. 2017, Berneira et al. 2020, Graeve et al. 2002, Santos et al. 2017, and Schmid et al. 2018
The identity as well as the relative proportion of the main FA contributors in Ochrophyta 
in our study were extremely similar to those reported in the literature (Table 2.7). The 
Ochrophyta in our study as well as those reported in the literature (Graeve et al. 2002, Aumack et
al. 2017, Berneira et al. 2020) contained high proportions of the same set of FA: PUFAs 20:4ω6 
(19.8% in our study vs. 16.0% in the literature) and 20:5ω3 (18.5% vs. 15.7%), the SAFA 16:0 
(12.6% vs. 15.2%) as well as the PUFAs 18:4ω3 (11.8% vs. 12.0%) and 18:1ω9 (10.7% vs. 
10.5%).  In part, this consistency may be based on the large overlap in species between the 
literature and our study, including six out of the eight total Ochrophyta we investigated (see 
Table 2.4). However, we added not only two new species (Cystosphaera jacquinotii and 
Microzonia australe) in our investigation but these species belong to different orders (Fucales 
and Syringodermatales, respectively) that have not been included in any previous analysis. 
Despite minor differences in the rank order of the main contributing FAP adge e1 pending on which 
species set is included (Table 2.7), the very high consistency we found in FA composition even 
after adding representatives of two new orders confirms the conservative FA makeup of 
Ochrophyta. This does not only apply to the Antarctic Ochrophyta but is typical of the main FA 
35
composition of Ochrophyta in general from a variety of regions and climate regimes (e.g., 
Khotimchenko 1998, Khotimchenko et al. 2002, Dalsgaard et al. 2003, Sohrabipour 2019 and 
references therein).
Chlorophyta are the least diverse macroalgal division in the Antarctic, with many species 
only occurring at lower latitudes along the WAP (Oliveira et al. 2020). Our study only contained 
one green algal species, Lambia antarctica, which was only previously investigated for its FA 
composition by Graeve et al. (2002). FA composition for this species matched well for some of 
the main FA contributors in our study, such as the SAFA 16:0 (19.8% in Graeve et al. 2002 and 
20.3% in this study). It matched reasonably well for 18:3ω3 (23.7% in Graeve et al. 2002 and 
31.6% in this study), but it differed considerably for 18:2ω6, which contributed nearly a quarter 
(22.3%) of FAs in Graeve et al (2002), but only 6.3% in our analysis. Conversely, 20:5ω3 was 
the third-largest contributing FA in L. antarctica in our study while this FA was only represented
with 5.0% in Graeve et al. (2002). L. antarctica is a siphonous, unicellular species, so it is 
possible that cell content is lost to various degrees when damaged during collection, and the 
thallus also disintegrates rapidly. This may contribute to some of the differences in FA profiles 
we saw to the previously published record.     
Marine macroalgae are a polyphyletic group, where the major divisions differ in many of 
their cellular and molecular properties (Chapman et al. 2012, Belghit et al. 2017). This explains 
the consistency of differences in major FA profiles among the three main divisions, which has 
been reported numerous times for macroalgae worldwide (e.g., Dunstan et al. 2005, Hanson et al.
2010, Galloway et al. 2012, Kumari et al. 2013). This strong divisional grouping has also been 
observed previously for Antarctic macroalgae (Graeve et al. 2002) and our addition of 14 newly 
36
profiled species supported this finding. Some previous studies have documented similar 
separations on the order and even the family level for non-Antarctic species (e.g., Galloway et al.
2012, Kumari et al. 2013). Here, we documented, for the first time, that Antarctic macroalgal FA
profiles are similarly conservative on the order and family levels.  
A subset of seven FAs described patterns of phylogenetic differentiation equally well if 
not more distinctly than the full suite of 40 FAs. The FAs that were driving differences among 
macroalgal divisions were primarily EFAs, including 20:5ω3, 20:4ω6, 18:4ω3, 18:1ω9, 18:3ω3, 
and 20:3ω3. Only one SAFA (16:0) contributed strongly to these patterns. A similar observation 
was made for Northwest Pacific macroalgae, where robust phylogenetic separation was also 
observed when only applying EFA (Galloway et al. 2012). EFAs are typically used as key 
markers in food web analysis as they cannot be synthesized in most higher consumers (Budge et 
al. 2006, Brett et al. 2016). These EFAs also are routinely recorded in the published literature of 
macroalgal FAs, allowing the published profiles to be ‘mined’ for applications in trophic 
analyses. A caveat of this approach may be that, despite the overwhelming consistency reported 
in the literature for characteristic FA profiles of macroalgal divisions, there are reports that 
macroalgal FAs can vary seasonally in content and relative proportions, which may make taxon-
specific profiles more difficult to resolve (Hernández-Carmona et al. 2009, Dethier et al. 2013, 
Barbosa et al. 2020).  However, macroalgae living in narrow temperature windows may be 
limited in their ability to modulate their FA profiles (Barkina et al. 2019), suggesting that 
Antarctic macroalgal FA profiles, especially EFA, are likely reasonably consistent over time. As 
FA analyses can be labor and time intensive, a targeted FA approach, including relying on 
37
published macroalgal FA values, could allow a broader application of Antarctic macroalgal FAs 
in food web studies.
When solely considering SI and C:N ratios, macroalgal species resolved into three major 
groupings that did not fully align with phylogenetic groups or match those obtained based on FA 
profiles. Instead, groupings seemed to be driven more by physiological properties of the species. 
Specifically, δ13C values are strongly influenced by the carbon uptake mechanisms of the 
macroalgal species, resulting in a separation of macroalgal species into two groups based on their
δ13C values. Despite generally employing C3 photosynthesis, some marine florideophyte 
Rhodophyta lack pyrenoids and carbon-concentrating mechanisms (CCMs), rendering them 
highly depleted in 13C, typically resulting in δ13C values of -30‰ or lower (Raven et al. 2002, 
2005, 2012). Some Antarctic red algae, however, can have low δ13C values despite being able to 
use bicarbonate and having CCMs (Beardall & Roberts 1999). The largest macroalgal grouping 
in our analysis were species with such low δ13C values, suggesting that underlying physiological 
differences in photosynthetic processes drive this separation. While the occurrence of 
macroalgae with low δ13C values is typically rare, it was a fairly widespread phenomenon in our 
dataset. This is confirmed by similar SI records in previous studies of several of the same 
Antarctic macroalgal species as examined in this study (e.g., Fischer & Wiencke 1992, Dunton 
2001). Interestingly, one Ochrophyta (D. anceps) and the sole Chlorophyta in our study (L. 
antarctica) also grouped with this low δ13C value group. However, we are not aware of CCMs in 
marine brown or green algae, suggesting there may be other physiological processes at play. 
Accordingly, the δ13C values also were the strongest driver separating algal species with our 
PCA based on SI data and C:N ratios.
38
Some of the macroalgal species formed a lose grouping based on high δ15N values. Of 
particular note is P. plocamiestris, which had high mean δ15N values >6‰. The δ15N values of 
macroalgae can be driven by environmental conditions, such as inorganic nitrogen availability in 
the surrounding waters (Viana & Bode 2013, Lemesle et al. 2016) or selective release of nitrogen
during photorespiration (Kim et al. 2013). The former is unlikely to be a driving factor in 
Antarctic waters as anthropogenic influences of eutrophication are not present on larger spatial 
scales, and point sources such as guano from penguin or other seabird colonies that could be 
implicated in the isotope values of marine organisms (Rossi et al. 2019) were not present in our 
sampling locations. High tissue nitrogen SI values have been associated with protein breakdown 
during photorespiration in shallow-water macroalgae, especially during emersion in intertidal 
algae (Kim et al. 2013). While we did not collect P. plocamiestris intertidally, it is a very 
shallow-water species that could experience periodic emergence, which could drive its very high 
δ15N value.    
We also included the C:N ratio of macroalgae in the ordination based on SI tracers as an 
added dimension, a not uncommon approach when applying dietary mixing models including 
primary producer sources (e.g., Wen et al. 2016, Brett et al. 2016), although it should be noted 
that the C:N ratio itself is not a dietary tracer. But briefly, C:N ratios were generally low, 
reflecting high tissue nitrogen concentrations driven by the nitrogen-replete waters of coastal 
Antarctica (Grotti et al. 2001). C:N ratios in our study were lower in Rhodophyta than 
Ochrophyta, confirming earlier findings by Weykam et al. (1996), although differences in thallus
structure have also been implicated in driving C:N ratios in Antarctic macroalgae and likely 
contributed to the variability we observed in the C:N ratios in the Rhodophyta. The higher C:N 
39
ratios in the large Ochrophyta are indicative of high levels of structural cell components (cell 
walls, phlorotannins, e.g., Iken et al. 2007). Thus, C:N ratios added to the divisional separation 
of macroalgae when combined in multivariate ordinations with the stable carbon and nitrogen 
isotope markers.     
The underlying processes driving macroalgal FAs versus SI values among macroalgal 
species are quite different, with FAs primarily driven by metabolic pathways rooted in 
phylogenetic identity, while stable isotope values are usually more influenced by environmental 
factors such as light, temperature and the ability to sequester inorganic carbon (Wiencke & 
Fischer 1990, Fischer & Wiencke 1992, Dalsgaard et al. 2003, Guest et al. 2010, Mackey et al. 
2015). In this context, it is of note that we did not find any spatial trends either in FA profiles or 
SI values of select macroalgal taxa across an environmental gradient spanning >3 degrees 
latitude and a range of sea ice cover along the WAP (Galloway and Iken, unpublished data). This
indicates that the data provided here for both biomarkers are representative of the taxa in general.
We, therefore, expected the combination of the two biomarkers to add to the ability to separate 
macroalgal groupings, ultimately for the purpose of tracing food web connections. This has been 
increasingly suggested or applied in several other systems across the world (e.g., Hanson et al. 
2010, Kelly & Scheibling 2012, Dethier et al. 2013), where the combination of multiple 
biomarkers aided in finer trophic resolution of benthic primary producers. We found that the 
combination of FA profiles with SI markers and C:N ratios only slightly improved the separation
of macroalgal species or groupings in multivariate space.  This may be explained by the much 
larger number of FA compared to isotope markers. It is possible that the more nuanced SI 
40
separation of macroalgae based on physiological processes is overwhelmed by the strong 
phylogenetic origins of FA profiles that also underlie some of the SI patterns.
This does not mean that a combination of FA and stable isotope biomarkers would not be 
useful in Antarctic coastal food web studies, and we propose the benefit of a combined approach 
for several reasons. For one, SI are commonly used to analyze trophic connections (Peterson & 
Fry 1987, Layman et al. 2007), and while there is still uncertainty about fractionation and trophic
enrichment (Post 2002), our understanding of SI in food webs is probably greater than that of 
FA. There is still a need for controlled experimental work to better understand FA uptake and 
trophic transfer (Galloway & Budge 2020). An example of the usefulness of a combined 
approach has been shown for Antarctic amphipods feeding on macro- and microalgae, where 
general food sources identified by SI data were confirmed and refined by FA analysis (Aumack 
et al. 2017). Also, the different underlying processes driving FA and SI in macroalgae can be 
particularly useful when aiming to differentiate the contribution of phylogenetically close 
macroalgal species to a consumer. Based on FA, those species would likely be difficult to trace 
but if they differ profoundly in their SI composition, the combination of the two biomarkers 
could prove very useful. For example, the two red algal species, Phyllophora antarctica and 
Callophyllis atrosanguinea, were almost identical in their FA profiles, and they also are 
morphologically difficult to distinguish (Wiencke & Clayton 2002). Isotopically, however, these 
two species were vastly different, with P. antarctica having very low δ13C values of around -
37‰ compared to -23‰ in C. atrosanguinea in our study, similar to values reported previously 
(e.g., Norkko et al. 2004, Marconi et al. 2011 for genus-level value for Callophyllis). Therefore, 
using multiple lines of evidence in tracing macroalgal species in the coastal food web could 
41
prove especially useful. Therefore, comprehensive data set across a large number of macroalgae 
for both FAs and SI provided in the present work provides an important baseline for the future 
application of these tracers in Antarctic food web studies.
BRIDGE
In Chapter II I focused on the chemical ecology of a suite of nearshore 
macroalgae in Antarctica. In Chapter III I consider a suite of nearshore macroalgae along
the West Coast of North America and how these communities change across a seascape 
in conservation and adjacent comparison areas. 
42
CHAPTER III
MULTI-YEAR MONITORING OF MACROALGAL ABUNDANCE AND DIVERSITY WITHIN
THE OREGON MARINE RESERVE SYSTEM AND COMPARISON AREAS
From Whippo R, Fields SA, Aylesworth L (2022). SCUBA kelp report. Oregon 
Department of Fish and Wildlife Marine Resources Program. Newport, 
OR. 
https://ecologyreports.oregonmarinereserves.com/Data_Files/7.%20Collab
orations/SCUBA%20Algal%20Swath%20Report/
ODFW_markdown_algae.html
1. INTRODUCTION
The importance of large seaweeds, or macroalgae, in coastal ecosystems cannot be 
understated. Distributed worldwide, they form the base of many marine food webs, provide 
habitat for a diverse collection of species across many life stages, and comprise some of the most
productive marine ecosystems in our oceans  (Seitz et al. 2014, Krause-Jensen & Duarte 2016, 
Duarte 2017, Lefcheck et al. 2019). Macroalgae form mosaics of habitat above the tide line, as 
well as in deeper subtidal waters. In Oregon there are more than 50 common low intertidal and 
subtidal macroalgal species that are used as shelter or food by ecologically and commercially 
important species including rockfish, salmon, sea urchins, and abalone (Seitz et al. 2014, Krieg 
et al. 2019, Lefcheck et al. 2019).
While much research has described the diversity and ecology of intertidal macroalgae in 
Oregon, there has been relatively little investigation into subtidal macroalgal communities, 
43
which potentially comprise a large portion of the habitable coastline (Menge et al. 1993, 2005, 
Bracken & Nielsen 2004). This is due, in large part, to the logistical difficulties of accessing 
near-shore submerged habitat exposed to the open ocean. Assumptions about subtidal macroalgal
diversity in Oregon have therefore been primarily based on intertidal algal observations across 
the coast, which are much more robust and have been summarized in numerous places (Krieg et 
al. 2019). Prior to the formation of the Oregon Marine Reserves monitoring program in 2010, the
only data available on subtidal macroalgal communities came from disparate sources. These data
included approximately 50 SCUBA surveys by the Partnership for Interdisciplinary Studies of 
Coastal Oceans (PISCO) from 2001-2004 (Carr et al. 2020), a contracted benthic resources 
summary by The Port Orford Ocean Resources Team (POORT) and ODFW in 2009, and 
satellite imagery targeting canopy-forming kelps in Oregon collected semimonthly since 1984 
that has only recently been synthesized (Hamilton et al. 2020). The remote-sensing data 
synthesis in particular is useful as it provides a continuous record of kelp cover on the Oregon 
Coast for the last 35 years, however, it targets only the large canopy forming kelp Nereocystis 
luetkeana leaving many sub-canopy species unaccounted for. While these historical surveys 
capture snapshots of algal communities on a broad scale, they lack continuity in the case of the 
PISCO data, algal abundance in the POORT report (presence/absence only), and assemblage-
level accounting in the case of the satellite data. What is needed is a continuous, fine-scale multi-
species inventory of subtidal algal resources along the Oregon Coast.
Although global canopy kelps are in decline, there is evidence that Oregon’s canopy kelp 
communities have increased in area since 2014, or at least remained stable (Krumhansl et al. 
2016, Hamilton et al. 2020). Yet the state and trajectory of Oregon’s subtidal algal communities 
which are associated with these canopy kelps is unknown. Warming sea surface temperatures 
44
and increasing sea urchin populations have been linked with large-scale macroalgal declines on 
the West Coast (Rogers-Bennett & Catton 2019). This speaks to an urgency to characterize our 
local macroalgal populations, as Oregon is in the transition zone between recently decimated 
kelp populations to the south, and more stable populations to the north (Pfister et al. 2017, Beas-
Luna et al. 2020). In addition, macroalgal communities have been shown to have differential 
responses to upwelling depending on the strength and duration of the event (Hessing-Lewis & 
Hacker 2013). As patterns of upwelling along the Oregon coast continue to shift with climate 
change, it is vital to implement long-term, targeted monitoring of Oregon’s subtidal algal 
resources to inform the state of these valuable macroalgal communities and how they might 
respond (Rykaczewski et al. 2015).
The goal of the ODFW Marine Reserves Ecological Monitoring program is to provide 
such long-term monitoring to track changing nearshore communities over time. The reserves 
were put in place with the expectation of preserving biodiversity and resilience of the coastal 
ecosystem. While systems such as these have been effective for fish and invertebrate 
communities in many areas around the world, less is known about their effects on macroalgal 
assemblages (Halpern 2003, Molloy et al. 2009). Studies suggest that it may be difficult to detect
community-wide algal trends even over a decade or more of protection, although targeting focal 
species may provide more answers than simply considering total biomass or other gross 
indicators of production (Barrett et al. 2009, Medrano et al. 2020).
As a first step to implementing long-term monitoring of the marine reserves’ and 
associated comparison areas’ marine flora and fauna, ODFW contracted CA-based PISCO divers
to start benthic SCUBA surveys in 2010 and 2011 at the Redfish Rocks and Otter Rock Marine 
Reserves. These surveys included a targeted kelp swath survey – focused on commonly observed
45
kelp species (brown algae) - and a benthic habitat survey (uniform point count (UPC)) focused 
on broad structural groups of mostly red algae). After this initial survey effort, it was determined 
that contracting PISCO to conduct future surveys was not a feasible long-term strategy, so 
ODFW worked with PISCO and local partners to build a volunteer dive program in Oregon 
based on PISCO methods. Starting in 2013, the Oregon Marine Reserves (ORMR) volunteer 
dive team began collecting monitoring data targeting macroalgae, fish, and other associated 
invertebrates. The Marine Reserve Program and local partners have worked hard to sustain the 
pool of well-trained volunteer divers that currently collect data at four of the five marine reserve 
(Cape Perpetua has no hard bottom habitat in diveable depths).
The data summarized in this report are from the kelp swath surveys, for data on benthic 
habitat and cover from UPC surveys, please see the SCUBA habitat and cover appendices for the
Redfish Rocks, Otter Rock, Cascade Head and Cape Falcon Marine Reserves. The kelp swath 
surveys target several common taxa identified to genera or species level including: Alaria 
marginata, Costaria costata, Desmarestia sp., Saccharina latissima, Laminaria setchellii, 
Nereocystis luetkeana, Pleurophycus gardneri, and Pterygophora californica. Of these, the 
ODFW Marine Reserves program selected Nereocystis luetkeana and Pterygophora californica 
as focal species based on their ecological, economic or management importance. For more 
information please refer to the (Update) Methods Appendix detailing focal species selection. 
These data are a starting point to describe diversity, abundance, and distribution of subtidal 
Oregon kelps, however, they are limited in their descriptions due to the depths, spatial extent, 
and temporally irregular intervals at which they were conducted. 
1.1 Knowledge gaps 
46
Given the dearth of information available on the diversity and distribution of subtidal 
macroalgal communities along the Oregon Coast, there are several key knowledge gaps that can 
be prioritized to provide information as a starting place to create a clearer picture of the state of 
these communities. These gaps include: 
1. The abundance and distribution of key subtidal macroalgae. This includes canopy kelps 
as well as perennial sub-canopy species.
2. The stability of macroalgal abundance and distribution through time. Where do we find 
macroalgae, and are assemblages of species consistent across years?
3. Associations between fish, invertebrates, and macroalgal habitats. Do we see associated 
changes in animal communities associated with macroalgae? Do these associations 
change across space and through time?
4. The overall diversity of subtidal macroalgae. What is the overall diversity of all subtidal
macroalgae in Oregon?
5. The oceanographic, climatic, and ecological factors controlling macroalgal diversity, 
composition, and abundance. There are many potential drivers of macroalgal 
community state that are likely operating across multiple scales (Lamy et al. 2018; 
Conser and Shanks 2019). 
The first three gaps are being partially addressed by by the OR Marine Reserves 
monitoring program, though methods can be streamlined, augmented, and targeted from lessons 
learned since monitoring of the reserves began. The last two are opportunities for future work 
that could be addressed though various means, including ‘bio-blitz’-style survey efforts, DNA 
analysis, and modeling.
47
1.2 Research questions
 What kelp species have been observed at each marine reserve and comparison area?
 How have these kelp communities changed over time?
2. TAKEAWAYS
Here we present the major takeaways from our SCUBA kelp swath analyses across 
Oregon’s Marine Reserves. 
2.1 All targeted kelp swath species have been observed across the marine reserves, but their 
presence varied among sites and years. 
Over the course of the Marine Reserve Monitoring Program all major groups of kelps 
targeted–including ‘focal species’ (N. luetkeana, P. californica) and other species of interest 
(Desmarestia sp., C. costata, L. setchellii, S. latissima, and A. marginata)–were detected, 
although the detection of those species varied among marine reserves, comparison areas, and 
years. The highest total counts of kelps within a marine reserve totaled across all years were 
found at Otter Rock, followed by Redfish Rocks and Cascade Head. No kelps were detected at 
Cape Falcon. P. californica and A. marginata are found almost exclusively in the shallow 
subtidal <10 meters depth so it’s not surprising that they were not frequently detected at any of 
the sites. 
2.2 Monitoring data suggest declines over time, but likely attributable to methodological shifts, 
not true biological changes.
Summary data from both the Redfish Rocks Marine Reserve and Otter Rock Marine 
48
Reserve suggest precipitous kelp declines between 2011 and later survey years. However, the 
low detection rate of kelp in later years and generally shallower transects (half the depth at Otter 
Rock) in earlier years (2010-2011) do not allow us to rule out methodological differences 
causing the declines observed. In general, a decline was detected at Cascade Head as well, but 
the low counts of kelps, even in earlier years, do not allow us to draw any conclusions. No kelps 
were detected at Cape Falcon in any survey so trends cannot be determined. At the Otter Rock 
Marine Reserve, the early PISCO surveys counted 3,136 kelp observations over two years, 
whereas the later two survey years counted only 107 individuals. If we compare the early two 
years to the later two years, we noted a difference in average transect depth, where the 2010-
2011 transects were on average five meters shallower than the 2017-2019 transects. At Redfish 
Rocks Marine Reserve, we also observe very low detection rates of kelp in the later two years of 
surveys (2015, 2019), compared to the earlier years (2010, 2011). While the differences in depth 
between these years was not significant, safety concerns of diving in the open ocean limited 
access to kelp beds in these later years (L. Aylesworth, pers. comm). It is possible that the results
may reflect a genuine decline in N. luetkeana in both regions, though other regional data on N. 
luetkeana populations through 2018 suggests they are increasing or stable, indicating there may 
be methodological differences at play (Hamilton et al. 2020). Furthermore, there are no 
observations in the literature of any decline on the order of what is observed in the data, if these 
were indeed true biological declines. Macroalgal communities up until 2010 had been fairly 
stable in estuaries across Oregon, though macroalgae are sensitive to temperature and the marine 
heatwave of 2015 could have negatively impacted these populations (Hessing-Lewis & Hacker 
2013, Straub et al. 2019), this is not supported by other regional N. luetkeana data (Hamilton et 
al. 2020).
49
2.3 Adapting SCUBA surveys to open ocean conditions in Oregon limits our ability to survey 
prime subtidal kelp habitat. 
On the surface, it appears there was a drastic decline in kelp detection between initial 
survey efforts and subsequent survey years for all sites except Cape Falcon where no kelp was 
detected. However, with the transition from the PISCO dive team to an Oregon-based volunteer 
dive team, surveys were adjusted for safety concerns around live-boating in nearshore 
conditions. This resulted in requiring a down-line divers must use to descend/ascend, avoiding 
the shallowest target depth (5 m), and avoiding N. luetkeana beds. Algal communities are 
concentrated in the shallows to access light for photosynthesis; surveys in later years moved 
away from targeting the shallower 5 m depths at which we’d expect to find dense nearshore 
macroalgal communities along the Oregon coast. Of the remaining two target depths (12.5 and 
20 m) only one of those has the potential for algae communities as the photic zone is frequently 
shallower than 20 m during summer months (Small & Menzies 1981, Steneck & Dethier 1994). 
Safety concerns of diving at such shallow depths in exposed open ocean environments, and the 
requirement to maintain contact with a descent line limited access to prime subtidal algal habitats
at some sites occasionally, and at other sites entirely.
These safety restrictions led to targeting the edges of N. luetkeana beds or other hard 
bottom substrate without N. luetkeana beds at the Otter Rock and Redfish Rocks Marine 
Reserves. The implementation of monitoring efforts at the Cascade Head Marine Reserve (2014) 
and Cape Falcon Marine Reserve (2016) provided additional challenges, as these sites are not 
known to contain the rich N. luetkeana beds of the other two marine reserve sites, and targeting 
the edge of such beds was not possible. These factors combined with the irregular sampling 
50
intervals and randomly stratified transect locations make it difficult to draw any conclusions 
about the state of kelp populations over time.
For these reasons, the program is not currently able to detect change over time in kelp 
communities apart from generalized presence/absence metrics. These data should be considered 
snapshots of limited transects at a particular place in time, though they provide a starting place 
for future survey work. 
2.4 While abiotic characteristics are generally similar between reserves and comparison areas, 
more data is necessary to assess the suitability from a kelp perspective.
From the kelp monitoring data we cannot say whether or not the comparison areas are 
appropriate reference sites for assessing change over time at the marine reserves. The ODFW 
Marine Reserves program did select comparison areas based on similar abiotic conditions (depth 
range, size, habitat types, oceanographic conditions), but further data is required to determine if 
comparison areas are appropriate from a kelp perspective. 
2.5 While methodological shifts in kelp monitoring proved to be challenging for analyses, data 
provides a ‘snapshot’ of subtidal kelps in Oregon.
The challenges of our kelp data have been outlined above, but we were able to collect 
some baseline data that provides a snapshot of more rarely sampled kelps, especially subcanopy 
kelps. With the recommendations below our program aims to improve the consistency of our 
kelp data collection.
3. RECOMMENDATIONS
51
3.1 A move towards permanent transects for kelp surveys is needed to confidently detect future 
trends in brown algae with SCUBA surveys.
Algal communities can be ephemeral, and random stratified selection of survey sites, 
particularly those that encompass substrate that we would not expect to have associated 
macroalgal communities, may not be representative of the population dynamics at play. 
Therefore, the installation of several permanent transects along appropriate substrate (i.e. - 
bedrock, consolidated rock) in extant kelp habitat will allow temporal variation to be tracked. 
These can be supplemented occasionally by additional surveys for annual ‘snapshots’ of the 
community.
3.2 Eliminate 20 m kelp surveys, reconsider adding 5 m sites where appropriate.
Kelp communities are concentrated in the shallows to access light for photosynthesis. 
The depths of the surveys (5 m, 12.5 m, 20m) only capture two depth contours at which we’d 
expect to find dense nearshore macroalgal communities along the Oregon coast (5 m, 12.5 m), as
the photic zone is frequently shallower than 20 m during summer months (Small & Menzies 
1981, Steneck & Dethier 1994). To address this shortcoming, more shallow surveys should be 
conducted to access the likely range of canopy, and subcanopy kelp Additionally, future analysis 
could focus on 12.5 m data only to explore trends, with this depth-bin having the most overlap 
between PISCO and ODFW surveys. 
3.3 Conduct surveys during peak of kelp production season (July – August).
The lack of consistent sampling across years poses a challenge in detecting patterns of 
52
kelp community dynamics. Surveys along permanent transects should be conducted at a 
minimum of once per year, preferably at the peak of macroalgal production season (June-
August) as is standard in other macroalgal monitoring programs (Pfister et al. 2017, Byrnes & 
Reed 2018) to ensure more comparability from year to year and increases the likelihood of 
sampling the full kelp community.
3.4 Consider adding in UAV surveys at Redfish Rocks and Otter Rock Marine Reserves to 
improve tracking of the focal species, N. luetkeana.
Additional surveys of N. luetkeana, one of the ‘focal’ species identified by ODFW for 
monitoring, can also be tracked using unmanned aerial vehicle (UAV, aka drone) technology. 
Drones have become and increasingly affordable and reliable way to quantify canopy kelps such 
as N. luetkeana in nearshore areas (Tait et al. 2019, Thompson 2021). These remote sensing 
techniques has also been used to monitor many types of marine vegetation and can be 
accomplished by surface crew concurrently with SCUBA transects and other shore- or boat-
based surveys.
4.  METHODS
4.1 Field surveys
All dive surveys were conducted from a boat within the Oregon Marine Reserve System 
and in associated comparison areas. The overall survey design for the initial surveys consists of 
multiple survey sites on shallow (< 20m depth) rocky reef habitat within each reserve area and 
corresponding comparison sites outside of the reserve. Comparison areas outside of each reserve 
were chosen based on similar depths, habitat, size, oceanographic conditions and historical 
53
fishing pressure similar to each reserve. In 2013, PISCO methods and datasheets were modified 
to adjust for common Oregon species, frequently challenging sea states, and safety needs given 
the volunteer diver team (avoiding shallower depths, avoiding kelp bed locations, and requiring 
downlines).
Kelp swath surveys were conducted concurrently with invertebrate and benthic habitat 
and cover surveys where depths of 12.5 and 20 m were targeted . Once the site was located with 
GPS, a video lander was deployed from the boat with a buoyed downline. Divers would descend 
on the downline and use the lander as the anchor-point for their transect tapes (a required OSU 
safety requirement). A visibility check was conducted on each dive with a minimum acceptable 
visibility of 1 m. A 30 m transect was laid along contiguous rocky substrate following the depth 
contour of interest and following the natural curves of the substrate. Transect depths were kept 
within 1.5 m of the target depth and any changes in substrate type were noted during data 
collection. Transects did not overlap with other transects and the surveys were aborted if more 
than 5 continuous meters of sand were encountered. Divers noted all algal species of interest 
within 1 m on either side of the transect including:
 Nereocystis luetkeana (>1 m stipe length)
 Pterygophora californica (>30 cm stipe length)
 Pleurophycus gardneri (>30 cm stipe length)
 Laminaria setchellii / Saccharina latissima1 (>30 cm stipe length)
 Costaria costata (any size)
 Desmarestia sp. (any size)
 Alaria marginata (any size)
54
Algae were subsampled if encountered in excess of 30 individuals per 10 m-swath 
segment. Subsampling ended at the completion of each swath column (i.e., every 10m), and 
regular counting resumed.
4.2 Data analysis
All data used in this analysis were complete 30 meter transects (i.e. equal area surveyed), 
thus abundance is explored at as counts. Survey effort and basic abundance relationships among 
sites, years, and species were visualized in boxplots and tables. Higher-level statistical analysis 
proved difficult due to the sparse nature of the data (zero-rich dataset) and irregular sampling 
intervals (skipped years, unbalanced sampling effort). While simple statistical comparisons were 
made between dive parameters with ANOVA and post-hoc Tukey tests, any higher-level 
analyses with factors of interest (site, year, reserve vs. comparison area) could only be achieved 
by pooling data sets or comparing data that were not statistically robust, which was not a realistic
or informative line of inquiry. As such, descriptive visualizations were created to suggest trends 
and provide a starting point for future survey design refinement. Because we adapted original 
PISCO protocols for long-term use in Oregon nearshore water, we decided to test data collected 
with the original PISCO protocol against those collected with the modified Oregon Marine 
Reserves (ORMR) method. To determine if overall mean transect depth was different between 
the protocols among sites a multi-way ANOVA was run. Significant differences in site-specific 
depths were detected using a post-hoc Tukey test. All survey and species counts were visualized 
in R using the tidyverse, gt, and viridis packages (Wickham et al. 2019, Garnier et al. 2021, 
Iannone et al. 2022, R Core Team 2022).
55
5.  RESULTS
5.1 Data overview
Between 2010 and 2019 there were a total of 254 kelp swath survey transects completed 
in the Oregon Marine Reserves and associated comparison areas (Table 3.1). Kelp was counted 
on 102 of the 254 total transects (~40% overall detection rate) at every site except Cape Falcon, 
where no kelp was detected.
Reserves and their associated comparison areas were surveyed annually on a rotating 
basis. Three of the reserves have four years of survey data (Redfish Rocks (2010, 2011, 2015, 
2019); Otter Rock (2010, 2011, 2017, 2019); Cascade Head (2013, 2014, 2017, 2018)), and 
despite several efforts over multiple years, only one year of survey data is available for Cape 
Falcon (2017). Kelp transects within a single reserve and comparison area ranged from 1-31 in a 
single year, with an average of 20 transects conducted per site on years when it was surveyed.
5.2 Protocol comparison
Between 2010 and 2019 there were a total of 254 kelp swath survey transects completed 
in the Oregon Marine Reserves and associated comparison areas (Table 3.1). Kelp was counted 
on 102 of the 254 total transects (~40% overall detection rate) at every site except Cape Falcon, 
where no kelp was detected.
Reserves and their associated comparison areas were surveyed annually on a rotating 
basis. Three of the reserves have four years of survey data (Redfish Rocks (2010, 2011, 2015, 
2019); Otter Rock (2010, 2011, 2017, 2019); Cascade Head (2013, 2014, 2017, 2018)), and 
despite several efforts over multiple years, only one year of survey data is available for Cape 
Falcon (2017). Kelp transects within a single reserve and comparison area ranged from 1-31 in a 
56
single year, with an average of 20 transects conducted per site on years when it was surveyed.
Table 3.1: Number of SCUBA kelp swath transects per year across all regions including marine 
reserves and comparison areas completed as part of Marine Reserves Ecological Monitoring. All 
surveys conducted in 2010-2011 followed original PISCO dive protocols. Those from 2013 
onward followed a modified ORMR protocol.
Fig. 3.1: Average depth of kelp swath transects between ODFW and PISCO survey teams.
5.3 Data by site
5.3.1 REDFISH ROCKS
57
Fig. 3.2: Map of SCUBA transect locations at Redfish Rocks Marine Reserve.
58
Fig. 3.3: Map of SCUBA transect locations at Humbug Comparison Area.
59
A total of 75 kelp transects were conducted at Redfish Rocks, and Humbug Comparison 
Area over four years (2010, 2011, 2015, 2019; Table 3.2, Figure 3.4).
Table 3.2: Total number of SCUBA kelp swath transects conducted in Redfish Rocks Marine 
Reserve and Humbug Comparison Area between 2010 and 2019.
Fig. 3.4: Total SCUBA kelp swath surveys conducted by year in the Redfish Rocks Marine 
Reserve (dark blue), and Humbug Comparison Area (lighter blue).
Mean depths of PISCO and ORMR protocols differed between the marine reserve and 
60
comparison area, though none of these differences were statistically significant (Table 3.3).
Table 3.3: Mean depth in meters of SCUBA kelp swath transects conducted with PISCO 
protocols (2010, 2011) and modified OR (ORMR) protocols (2015, 2019). None of the depth 
comparisons between protocols or sites were statistically significant (post-hoc Tukey, p > 0.05).
Across all years and protocols the focal species N. luetkeana was the most 
abundant alga of all species surveyed within the Redfish Rocks Marine Reserve and in 
the Humbug Comparison Area (Figure 3.5). Focal species P. californica was detected 
also in three of the four survey years, along with associated species L. setchellii and P. 
gardneri. No kelp was detected inside or outside the reserve in 2019.
 Mean depth of transects did not differ significantly between the PISCO and ORMR 
surveys.
 The focal species N. luetkeana and P. californica, along with other common species P. 
gardneri and L. setchellii were detected in the marine reserve between 2010-2015, but 
not detected at all in 2019. Observations in the comparison area followed a similar 
pattern.
 Observed declines in N. luetkeana may be due to methodological differences between the
PISCO and ORMR surveys. Specifically the avoidance of N. luetkeana beds by ORMR 
due to safety considerations.
61
Fig. 3.5: Kelp species counts observed in the Redfish Marine Rocks Reserve and Humbug 
Comparison Area (combined) by year.
The focal species N. luetkeana was found in relatively high abundance within the Redfish
Rocks Marine Reserve in 2010 and 2011 along with associated species P. gardneri and L. 
setchellii, although L. setchellii was found less frequently in 2011 with a total decline in 
detection of 76% (Table 3.4, Figure 3.6). Focal species P. californica peaked in 2015 with 139 
individuals counted, but was not detected at all in the following survey year of 2019. In fact, no 
kelp was detected at Redfish Rocks Marine Reserve at all in 2019.
A total count of 1013 individual N. luetkeana were found in the Humbug Comparison 
Area in 2011, the second highest detection of N. luetkeana across all years and sites second only 
to Otter Rock Marine Reserve in the same year, and the highest count within a comparison area
62
Table 3.4: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Redfish Rocks Marine Reserve by year. Total counts are 
summed across 30 m transects.
 (Table 3.5, Figure 3.6). Algal detection was reduced greatly in 2015 and 2019, similar to the 
marine reserve, however none of the focal species P. californica was detected in Humbug 
Comparison Areas as it was in the marine reserve. Overall, kelp detection between the marine 
reserve and comparison areas tended to be species-specific in terms of abundance within and 
across years. The only consistent pattern was the drastic reduction in detection between 2010/11 
and 2015/19. The common species Desmarestia sp. was also detected in small quantities in 
2015/19.
________________________________
Table 3.5: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Humbug Comparison Area by year. Total counts are 
summed across 30 m transects.
63
Fig. 3.6: Kelp species counts observed in Redfish Rocks Marine Reserve and Humbug 
Comparison Area from SCUBA kelp swath surveys by year. Total counts are summed across 30 
m transects.
5.3.2 OTTER ROCK
________________________________
Fig. 3.7: Map of SCUBA transect locations at Otter Rock Marine Reserve.
64
65
A total of 63 transects were conducted at Otter Rock over four years of survey effort 
(2010, 2011, 2017, 2019; Table 3.6, Figure 3.8).
Table 3.6: Total number of SCUBA kelp swath transects conducted in Otter Rock Marine 
Reserve and Cape Foulweather Comparison Area between 2010 and 2019.
Fig. 3.8: Total SCUBA kelp swath surveys conducted by year in the Otter Rock Marine Reserve 
(dark blue), and Cape Foulweather Comparison Area (lighter blue).
66
Mean depths of PISCO and ODFW transects differed significantly among the Marine 
Reserve and Comparison Area transects. PISCO transects were reliably shallower than the 
ODFW transects with a mean difference of 5.57 m within the Marine Reserve (n = 41) and 7.05 
m in the Cape Foulweather Comparison Area (n = 22; Table 3.7).
Table 3.7: Mean depth of kelp surveys conducted at Otter Rock by PISCO (2010, 2011) and 
modified OR (ORMR) protocols (2017, 2019). Depths between teams were different in both the 
Marine Reserve and Comparison Area (post-hoc Tukey, ** p < 0.001, *** p < 0.0001).
Across all years the common species L. setchellii and S. latissima were the most 
consistently detected at Otter Rock, while focal species P. californica and N. luetkeana were the 
most abundant overall in 2010 and 2011 (Fig 8). This was also the only site where the alga 
Costaria costata was found inside or outside the Marine Reserve system.
 Transect depths were significantly shallower using the PISCO protocol (2010, 2011) by 
5.57 m and 7.05 m in the marine reserve and comparison area respectively.
 The Cape Foulweather Comparison Area was the only site among all the regions 
surveyed where the alga Costaria costata was detected (2011, 2019).
 Kelp detection was greater by an order of magnitude in the marine reserve than in the 
associated comparison area in 2011. This was driven primarily by N. luetkeana and P. 
californica. While detection remained higher within the marine reserve in 2017 and 2019,
overall counts were greatly reduced.
67
Fig. 3.9: Kelp species counts observed in the Otter Rock Marine Reserve and Cape Foulweather 
Comparison Area (combined) by year.
Inside the Otter Rock Marine Reserve the focal species P. californica was detected in 
every survey year. Likewise, the common alga P. gardneri was also detected each year, albeit in 
much lower quantities (Table 3.8, Figure 9). Within the marine reserve N. luetkeana was 
detected at its highest level with a count of 1702 individuals across 14 transects. The alga 
Desmarestia sp. was also detected in 2017.
________________________________
Table 3.8: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Otter Rock Marine Reserve by year. Total counts are 
summed across 30 m transects.
68
In the Cape Foulweather Comparison Area both N. luetkeana and P. californica were 
detected in 2011, though P. californica was detected at lower levels than in the marine reserve 
the same year (Table 3.9, Figure 3.10). Only a single observation of P. californica was made in 
2017 and 2019, and N. luetkeana was not detected at all after 2011. Other species including P. 
gardneri, Desmarestia sp., C. costata, L. setchellii, and S. latissima were detected at very low 
level sporadically across the 2011 and 2019 survey years. No kelp whatsoever was detected at 
Cape Foulweather Comparison Area in 2017.
Table 3.9: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Cape Foulweather Comparison Area by year. Total 
counts are summed across 30 m transects.
69
Fig. 3.10: Kelp species counts observed in Otter Rock Marine Reserve and Cape Foulweather 
Comparison Area from SCUBA algal swath surveys by year. Total counts are summed across 30 
m transects.
5.3.3 CASCADE HEAD
A total of 88 transects were conducted at Cascade Head over four years of sampling effort (2013,
2014, 2017, 2018; Table 3.10, Figure 3.11-3.14).
Cascade Head Marine Reserve and associated comparison areas were not surveyed with 
the original PISCO protocols as other sites in 2010 and 2011. To detect if there are mean 
differences in transect depths across years, ODFW transects were binned into either a 2013-2014
or 2017-2018 category. These binned year groups were tested for differences in the same way as
70
Fig. 3.11: Map of SCUBA transect locations at Cascade Head Marine Reserve.
71
Fig. 3.12: Map of SCUBA transect locations at Schooner Creek Comparison Area.
72
Fig. 3.13: Map of SCUBA transect locations at Cavalier Comparison Area.
73
Table 3.10: Total number of kelp transects conducted in Cascade Head Marine Reserve and 
Schooner Creek and Cavalier Comparison Areas between 2013 and 2018.
Fig. 3.14: Total SCUBA kelp swath surveys conducted by year in the Cascade Head Marine 
Reserve (dark blue), and comparison areas (lighter blue).
 the PISCO and ORMR protocols in other marine reserves and comparison areas. Mean depths of
ODFW transects differed significantly across years, but not across sites within years (Table 
74
3.11). The 2013-2014 transects were 3.98 m shallower on average than the 2017-2018 transects.
Table 3.11: Mean depth of kelp surveys conducted at Cascade Head by the Marine Reserve 
Research Dive Team (ORMR) binned into 2013-2014 and 2017-2018 groups. Mean depth was 
different between years, but not among sites (ANOVA, df = 1, F = 14.010, ** p < 0.001).
Across all years inside and outside the Cascade Head Marine Reserve the most common 
kelp detected were Desmarestia sp., L. setchellii, N. luetkeana, and P. gardneri (Figure 3.15). 
Both focal species (N luetkeana and P. californica) were only detected in 2017, and a gradual 
decrease in overall detection of all kelp can be seen between 2013 and 2018.
 Surveys were consistently shallower in 2013-2014 than in 2017-2018, though no clear 
pattern in kelp detection other than an overall decline in 2018 was apparent.
 The focal species N. luetkeana and P. californica were only detected in the region in 
2017 within the marine reserve.
 The alga Desmarestia sp. was detected at the highest rate across all years and regions 
within the Cavalier Comparison Area at 71 individuals counted in 2013.
Within the Cascade Head Marine Reserve the focal species N. luetkeana and P. 
californica were only detected in 2017, while the other algae P. gardneri and L. 
setchellii/S. latissima were only detected in small quantities in 2014/17 and 2017/18 
respectively (Table 3.12; Figure 3.16). N. luetkeana had the highest detection count at 45,
which was found on just two transects. No other kelp species were detected in any year.
75
Fig. 3.15: Kelp species counts observed in the Cascade Head Reserve and comparison areas 
(combined) by year.
The only species detected at the Cavalier and Schooner Creek Comparison Areas were P.
gardneri, Desmarestia sp., L. setchellii/S. latissima (Table 13, Figure 13). Detection rate were 
generally higher in 2013-14 with 150 total counts of kelp across all comparison areas, but only 2 
detections in 2017-18.
________________________________
Table 3.12: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Cascade Head Marine Reserve by year. Total counts are 
summed across 30 m transects.
76
Table 3.13: Abundance of focal species (N. luetkeana, P. californica) and other common kelp 
from SCUBA swath surveys within the Cavalier and Schooner Creek Comparison Areas by year.
Total counts are summed across 30 m transects.
77
Figure 3.16: Kelp species counts observed in Cascade Head Marine Reserve and associated 
comparison areas from SCUBA kelp swath surveys by year. Total counts are summed across 30 
m transects.
5.3.4 CAPE FALCON
A total of 28 transects were conducted at Cape Falcon over one year of sampling effort 
(2017; Table 14, Figure 15), despite multiple attempts over several years to reach the Marine 
Reserve and comparison areas. Mean depths of transects were not significantly different among 
the marine reserve and comparison area transects with an overall mean depth of 14.2 m (Table 
15).
78
Fig. 3.17: Map of SCUBA transect locations at Cape Falcon Marine Reserve.
79
Fig. 3.18: Map of SCUBA transect locations at Low Fishing Pressure Comparison Area.
80
Fig. 3.19: Map of SCUBA transect locations at Moderate Fishing Pressure Comparison Area.
81
Fig. 3.20: Map of SCUBA transect locations at High Fishing Pressure Comparison Area.
82
Table 3.14: Total number of SCUBA kelp swath transects conducted in Cape Falcon Marine 
Reserve and Low Fishing Pressure and Moderate Fishing Pressure Comparison Areas since 
surveys began.
Fig. 3.21: Total SCUBA kelp swath surveys conducted by year in the Cape Falcon Marine 
Reserve (dark blue), and associated comparison areas (lighter blue).
83
Table 3.15: Mean depth of kelp surveys conducted at Cape Falcon by the Marine Reserve 
Research Dive Team (ORMR) separated into the Marine Reserve and associated comparison 
areas. Mean depth was not different among the marine reserve and comparison areas (post-hoc 
Tukey Test, p > 0.05).
 Due to the logistical difficulties of accessing Cape Falcon by SCUBA, kelp swath 
surveys were only conducted in 2017 within the marine reserve and at associated 
comparison areas.
 Although 28 SCUBA kelp swath surveys were conducted in the region, no kelp was 
detected on any transect.
BRIDGE
This Chapter addressed the status of macroalgae abundance and diversity within and near
a marine reserve system. In the next chapter, I focus on an important generalist predator in kelp 
forests that plays an important role in kelp forest health, the sunflower sea star Pycnopodia 
helianthoides.
84
CHAPTER IV
FEEDING PREFERENCES OF THE SUNFLOWER SEA STAR PYCNOPODIA HELIANTHOIDES
This study contains unpublished coauthored material that was conceived by me with conceptual 
input from Aaron Galloway. Field collections were made by myself and coauthor Kindall Murie. I was 
responsible for all experimental lab work with support from Kindall Murie. All statistical analyses were 
done by myself, and I wrote the manuscript with editorial input from all authors. 
1. INTRODUCTION
Predator-prey interactions are a cornerstone of ecosystem function and can control a suite
of processes including primary productivity (Estes & Duggins 1995), nutrient cycling (Schmitz 
et al. 2010), and patterns of biodiversity (Ellingsen et al. 2015). These interactions are modulated
by the environment in which they occur (Draper & Weissburg 2019), as well as through the 
specific behaviors and physiological states of the species involved in the interaction (Schmitz 
2017). The reciprocal responses of prey-to-predator and predator-to-prey can change outcomes 
of these interactions through various mechanisms including predator avoidance strategies (e.g., 
crypsis, escape; Heithaus et al. 2009, Spyksma et al. 2017) or prey preferences, including prey 
switching (Murdoch 1969). Prey switching is a common strategy predicted by optimum foraging 
theory when a predator pursues atypical prey because their preferred prey item is scarce or 
absent in the environment, or when more desirable prey become available (Prokopenko et al. 
2023). Depending on the capacity of a particular predator to switch prey, prey switching may be 
easier than other strategies such as long-range foraging, or alteration of hunting methods to 
obtain rare or evasive prey (Vallina et al. 2014). In habitats in which scarcity of potential prey is 
85
not a factor, other properties of available prey such as handling time, relative nutritional value, or
ease of capture dictate the choices predators make to pursue the prey item (Pyke et al. 1977, 
Ostfeld 1982).
The choices that predators make can dictate environmental outcomes such as the 
formation of alternate stable states (Beisner et al. 2003) or changes in community diversity 
(Douglass et al. 2008), which are often influenced by the relative abundance of prey in the 
environment (Hughes & Croy 1993). Other factors including predator-prey size ratios (Kalinkat 
et al. 2011), spatial overlap (Kempf et al. 2008), and ontogeny (i.e., juvenile versus adult; Elliott 
2006) can also play important roles in prey selection, and thus, the resulting ecological effects. In
addition, geographically widely distributed predator species may display different diet 
preferences based upon local (i.e., 10’s of km) environmental factors that may not be 
generalizable to their entire range (Byrne et al. 2019).
In kelp forests, which are highly productive habitats that are biodiverse and contain large 
quantities of biomass compared to many adjacent systems, predator-prey interactions play 
important roles in maintaining community structure (Dayton 1985, Sala & Graham 2002). 
Generalist predators in kelp forests have a wide range of prey items among which they can 
choose or potentially switch (Steneck et al. 2002), from small mobile fishes and invertebrates to 
encrusting organisms. Prey species identity can be very important when considering how 
predation can influence a community, as particular prey species play different roles in the 
ecosystem across spatial scales, from centimeters to kilometers. For example, prey species such 
as chitons grazing on encrusting species clear occupied space on small (cm) scales and allow the 
settlement of new larvae (Elahi & Sebens 2013). On larger (km) spatial scales, some grazers 
such as sea urchins can denude entire kelp forests (Hamilton & Caselle 2015). The consumption 
86
of prey with different ecological functions can, therefore, result in community-wide effects that 
act across spatial scales, even resulting in trophic cascades, dependent upon what function their 
prey fulfills within the ecosystem (Ripple et al. 2016).
The sunflower seastar Pycnopodia helianthoides (hereafter Pycnopodia), is a widely 
distributed generalist consumer in North Pacific kelp forests and has been implicated as a 
primary predator of multiple taxa including bivalves and echinoderms (Mauzey et al. 1968, 
Duggins 1983). They have often been cited as a major predator of grazing sea urchins, including 
purple (Strongylocentrotus purpuratus, Bonaviri et al. 2017), green (Strongylocentrotus 
droebachiensis, Schultz et al. 2016), and red (Mesocentrotus franciscanus, Nishizaki & 
Ackerman 2007) urchins, all of which are, in turn, primary grazers of kelp. This Pycnopodia-
urchin-kelp relationship is becoming more and more recognized as a potentially major driver of 
kelp forest health (Galloway et al. 2023), particularly since Pycnopodia populations were 
dramatically reduced, and in some places completely extirpated, by seastar wasting disease from 
2012-2015 (Harvell et al. 2019, Hamilton et al. 2021). Pycnopodia’s decline was concurrent with
an increase in grazing urchin populations, likely from release of predation pressure, and a 
subsequent decline in kelp forest cover in many areas across the West Coast of North America 
(Burt et al. 2018, Rogers-Bennett & Catton 2019). The resulting mosaic habitat of remaining 
kelp forest patches and urchin barrens is marked by an overall reduction of biomass in formerly-
productive kelp communities (Rogers-Bennett & Catton 2019). Starved urchins within the 
barrens that are interspersed among the remaining kelp maintain this alternate stable state by 
continuously grazing on newly recruiting kelp, inhibiting the growth and development of kelp 
and other macroalgae (Spindel et al. 2021). The biodiversity within urchin barrens also is lower 
than kelp forests, which further decreases ecosystem functioning in barrens (Gabara et al. 2021). 
87
Therefore, grazing control of urchins through generalist predators such as Pycnopodia could be 
critical in maintaining important ecosystem functions.
Direct consumptive effects of Pycnopodia on the urchins in Pacific Northwest kelp 
forests are an important pathway for grazer control (Galloway et al. 2023). However, this grazer 
control depends directly on Pycnopodia prey selection. There are relatively few studies that 
describe the ‘wild diet’ of Pycnopodia or explicitly test their preference of particular prey types 
in situ, and the majority of field studies quantifying the effect of Pycnopodia on urchins is 
largely correlative in nature (e.g., Schultz et al. 2016, Bonaviri et al. 2017, Burt et al. 2018, 
Rogers-Bennett & Catton 2019). Qualitative observations of Pycnopodia feeding in the wild 
have indicated that they actively prey on urchins, bivalves, and gastropods, while also 
scavenging on fish, other seastars, seabirds, octopus, large crabs (e.g., Cancer productus, 
Galloway personal observation) and seals (Phoca vitulina, Whippo personal observation). This 
wide range of diets has resulted in Pycnopodia most frequently being identified as ‘generalist 
predators’ (Duggins 1983) and ‘facultative scavengers’ (Brewer & Konar 2005). Of course, these
are potentially complimentary terms, and it is likely that Pycnopodia plays different trophic roles
depending upon the depth at which they are found (i.e., intertidal versus subtidal), geography 
(i.e., local prey spectrum), and season. It is also likely that Pycnopodia, like other sea star 
species, can specialize in particular prey types depending upon various physiological and 
ecological drivers including ontogeny (Manzur et al. 2010), prey density (McClintock & 
Lawrence 1985), and intra- and interspecific competition (Gaymer et al. 2002, Storero et al. 
2020). The latter is of particular interest as competitor identity is often linked to habitat type 
(e.g., predators that specialize on hard-bottom versus soft-bottom prey). Pycnopodia are known 
88
to regularly cross habitat-type boundaries, particularly between hard- and soft-bottom substrates 
(Mauzey et al. 1968, Shivji et al. 1983, Sloan & Robinson 1983). However, the influence of 
ecological drivers such as competition and habitat type on Pycnopodia feeding preference and 
how their preferences may change as their habitats are altered (e.g., kelp forest to urchin barren) 
and the suite of competitors shift (e.g., repatriation of sea otters) is unknown.
To address some of these uncertainties, particularly related to the ecology of kelp forests, 
the feeding rates and relative preference of Pycnopodia on sea urchin species and other common 
benthic prey species from rocky nearshore habitats needs to be determined. This is especially 
true as the roles that Pycnopodia’s potential prey play in kelp forests vary, and Pycnopodia prey 
preferences may have dramatic, cascading consequences for the kelp forests they inhabit. 
Therefore, our goal for this study was to 1) review our state-of-knowledge about the Pycnopodia 
‘wild-diet’ based on previously published research; 2) use a meta-analysis approach to detect 
quantitative generalities about Pycnopodia diets across part of their geographic range; and 3) 
directly determine Pycnopodia prey choice preferences and feeding rate in a cafeteria-style 
experiment using common potential prey items.
2. METHODS
2.1 Meta-analysis
Published records on Pycnopodia feeding, prey spectrum and prey preference were 
searched on Google Scholar and Web of Science using the keyword ‘Pycnopodia helianthoides’. 
Published peer-reviewed and non-peer reviewed (thesis) papers were selected based on the 
following criteria: the study described stomach contents of Pycnopodia observed or caught in the
89
field; stomach contents were quantified in some way (count, percent); at least phylum-level 
taxonomy was used to identify the consumed organisms; and the data were presented in a 
systematic way (i.e., a table, list, figure, etc.). Data from each study were extracted by hand and 
collated, either through extraction of raw values presented in the text, or estimation of values 
through visual analysis of figures using open-source imaging software (The GIMP Development 
Team, 2019). These data on Pycnopodia prey were then separated by geographic area and 
location of observation (subtidal, intertidal), then summarized and visualized using R and the 
tidyverse, viridis, maps, scatterpie, and ggpubr packages (Wickham et al. 2019, Garnier et al. 
2021, Becker et al. 2022, R Core Team 2022, Yu 2022, Kassambara 2023). Prey categories were 
combined into broad phylogenetic groupings that would encompass most of the study results and
allow comparative interpretation. Diet across studies was further analyzed for major prey 
contributors to diet and trends in observations across depths (see statistical analyses below).
2.2 Cafeteria experiment
Pycnopodia for the prey choice experiment were collected from the subtidal benthos at 
sites in the San Juan Archipelago through a permitting arrangement between the Friday Harbor 
Labs (FHL) and Washington State (Washington State House Bill 68, R.C.W.28.77.230, 1969 
Revision R.C.W.28B.20.320) between June 26, 2022 and August 17, 2022. The collected 
Pycnopodia were held in flow-through sea water tanks (60 cm x 90 cm x 30 cm, WHD) prior to 
experimentation. As collections were opportunistic in nature due to the scarcity of Pycnopodia, 
holding times ranged between 2-43 days before experiments were run. During the holding time, 
Pycnopodia were fed a daily maintenance diet of mussels and were handled according to 
protocols described in Hodin et al. (2021). All Pycnopodia were returned to their point of origin 
90
after the experiment was complete to minimize the impact that extraction could have on their 
local populations.
For the experiment, Pycnopodia were placed in outdoor flow-through sea tables (circular:
50 cm depth x 104 cm diameter) covered in black plastic sheeting to reduce light, and covered 
with an opaque lid. Temperature and salinity were monitored throughout each experimental trial.
As tank numbers were limited, no more than three Pycnopodia could be tested at any one time, 
each in its own separate tank. No Pycnopodia was tested more than once for prey preferences, 
and all tanks were drained, scrubbed, rinsed with fresh water, and left to dry between uses. A 
total of 11 Pycnopodia with a mean diameter of 48.8 ± 4.7 cm were used in feeding trials. We 
note that larger sample sizes of individual stars might have been preferable but were not possible 
due to the endangered status and rarity of these animals in the wild (Gravem et al. 2021).
Putative prey species were chosen to include three common shallow subtidal urchin 
species found around the San Juan Islands (Strongylocentrotus purpuratus, S. droebachiensis, 
Mesocentrotus franciscanus), an additional non-urchin echinoderm (the sea cucumber 
Apostichopus californianus), and two molluscs including mussels (Mytilus spp.) and a chiton 
(Cryptochiton stelleri). Prey items were collected from the FHL surroundings throughout the 
experiments to maintain sufficient supply of each taxon. For each experimental run, two 
individuals of each prey species were measured for body size and placed in the tank with one 
Pycnopodia to allow direct interaction. Prey consumption was checked and recorded at least 
three times per day over a ten-day experimental period, as well as opportunistically throughout 
the experiment. All prey items were replenished as soon as an ‘envelopment’ event (defined 
below) occurred to ensure at least two of each prey species were available in the tank at all times.
Envelopment was defined as the holding or covering of the prey item by the Pycnopodia so that 
91
the prey item was no longer visible within the tank. Egestion was defined as the expulsion of 
non-digestible materials (tests, shells), or thick mucus (sea cucumber) by the Pycnopodia after a 
feeding event. Envelopment or egestion that occurred overnight was set as occurring at midnight 
of the current observation day. Total number, identity, and size of prey items consumed were 
calculated over each experimental period for each Pycnopodia trial, with egestion time set as the 
completion of a consumption event. Pycnopodia prey items were removed after ten days, but 
Pycnopodia were held and observed into the eleventh day to ensure that all egestion events were 
recorded.
2.3 Statistical analyses
The prey taxa documented across all meta-analysis studies were reduced to eight 
categories for analysis (Table 4.1). We developed a relative measure of importance for each of 
those categories across all studies, calculated separately for subtidal and intertidal observations 
as a ‘weighted contribution factor’ (WCF):
pi ⋅ log10 [N j ]
WCF = 
( pi ⋅ log10 [N j ])max
Where pi = percent of each of the observed prey items in the diet for study i (separated into 
intertidal/subtidal, if applicable), and Nj = number of Pycnopodia used to calculate those 
observations in study i (also separated into intertidal/subtidal, if applicable). The values were 
then constrained between 0 and 1 by dividing by the maximum value in the entire dataset for 
92
ease of interpretation and to reduce the out-sized effect of outliers. Differences among WCFs 
were tested across all data, and separately for the intertidal and subtidal data with a Kruskall-
Wallis test, a non-parametric analysis that is robust to non-normal distributions and 
heterogeneous variance in JMP v17.0.0 (SAS Institute Inc., Cary, NC, USA). Post-hoc Wilcoxon
comparisons were made to determine food items that were driving differences in the observed 
diets. 
To determine feeding preferences of Pycnopodia in cafeteria experiments we used 
Rodgers’ preference index (Rodgers 1990), which accounts for order, rate, and total amount of 
each food item consumed, and is derived as the standardized area under the curve (AUC) of 
cumulative proportion eaten through time. This metric is considered appropriate for cafeteria-
style experiments and provides a value to indicate preferences from 0 (avoided) to 1 (preferred; 
Gasperini et al. 2018). Differences in preference between prey types and variation in AUC were 
tested with a Kruskall-Wallis test and Dunn’s post-hoc test, respectively. An additional analysis 
of the effect of urchin size on feeding choices was conducted using a linear mixed effects model 
(lme).
To further compare prey choices among individual Pycnopodia, the Shannon Diversity 
Index and Pielou’s Evenness were measured for each Pycnopodia prey consumption. These 
analyses provide a single value for the diet of each Pycnopodia, measuring the diversity of prey 
items in the diet (Shannon diversity), and the distribution of prey consumed (evenness) (Cramer 
et al. 2020). These measures provide additional information as they take into account items that 
were offered, but not consumed, and evenness provides the additional power of the null 
expectation that all prey were consumed equally (diversity = 1.61, evenness = 1).
93
The Pycnopodia used in the cafeteria experiment were collected from five distinct areas, 
ranging from high-flow kelp forests with hard-bottom substrate to low-flow pilings with soft-
bottom substrate. To determine if the location from which the Pycnopodia were collected had 
any meaningful effect on patterns of prey consumption, all Pycnopodia consumption patterns 
were tested against each other with PERMANOVA using location as a factor (Oksanen et al. 
2018).
Table 4.1: List of all prey items identified within each study and the corresponding category it 
was assigned to for analysis in this meta-analysis. Species name were not updated with current 
taxonomy due to uncertainty around acutal identity of species, however, categories were broad 
enough to account for any species-level changes. 
94
Table 4.1 (cont.)
95
3. RESULTS
3.1 Meta-analysis
Only six published datasets were identified that met the criteria for the literature search 
(Table 4.2). These studies spanned 24° of latitude along the West Coast of North America from 
Prince William Sound, Alaska in the north, to Pacific Grove, California in the south. The studies 
were all done within a 15-year period from 1968-1983, and sample sizes (number of Pycnopodia 
investigated) ranged widely from 7-425 individuals (Figure 4.1A).
A total of 114 prey taxa were reported across all studies (Table 4.1). Across all regions, 
bivalves and urchins tended to be the primary observed prey items, in both the intertidal (bivalve:
WCF = 0.46 ± 0.50; urchin: WCF = 0.35 ± 0.40) and subtidal (bivalve: WCF = 0.24 ± 0.25; 
urchin: WCF = 0.21 ± 0.27). Gastropods also were a common although rarely dominant prey 
item. There was, however, much variation among studies (Figure 4.1B). Differences among all 
prey types were statistically significant only when considering WCF across all locations and 
depths (Kruskall-Wallis, Chi-square(1, 7) = 19.08, p = 0.008), with bivalves and gastropods 
found to be statistically more abundant than other prey items (Post-hoc Wilcox, p < 0.05). There 
were no obvious strong latitudinal trends in patterns of consumption, although the majority of 
bivalve consumption tended to occur in the more northern region and very little urchin 
consumption was observed at the southern most site. The dataset was also marked by large 
spatial gaps between most sites.
96
Figure 4.1: Summary of Pycnopodia wild-diet studies from the literature (A) and relative prevalence of diet contributions across studies (B). 
Includes six studies published between 1968 and 1983. Prey types are grouped into relevant categories and identified by location of data 
collection (intertidal – green circle/header, subtidal – teal circle/header). N of Pycnopodia used are listed as numbers next to each pie, 
superscript letter denotes study of origin (a = Mauzey et al. 1968, b = Duggins 1983, c = Shivji et al. 1983, d = Paul & Feder 1975, e = 
Herrlinger 1983, f = Sloan & Robinson 1983). Total diet is summarized as mean percentage across all Pycnopodia observed from each 
sampling unit within each study. Weighted contribution of prey items to diet across studies is shown as mean and standard error (B).  
Table 4.2: Publications used in meta-analysis of Pycnopodia wild-diet. N is derived as number of
Pycnopodia observed feeding within each study area. As some studies made observations across 
the seascape, there may be multiple N within each. Unit is the type of measurement extracted 
from the study (percent of gut contents, count of individual items in gut contents). 
3.2 Cafeteria experiment
 In the lab experiments, Pycnopodia demonstrated high individual-level variation in 
feeding quantity, ranging from only a single prey item consumed, to 27 consumed items over the 
course of a ten-day trial (Figure 4.2). They also showed individual variability in prey type 
selection, consuming ≤ 2 types of prey (n = 2 Pycnopodia), 3 types of prey (n = 5), or 4 types of 
prey (n = 4) (Figure 4.2). These selections were not consistent by prey species; species that were 
consumed most often when summed over all trials were green urchins (n = 49), purple urchins (n
= 44), and mussels (n = 41). Red urchins were consumed 17 times and only a single cucumber 
was consumed. Chitons were never consumed in any of the feeding trials, though many 
developed scarring on their dorsal sides throughout the feeding trials, likely caused by attempted 
feeding by the Pycnopodia. On average, Pycnopodia consumed 0.41 ± 0.36 green urchins day-1, 
0.36 ± 0.41 purple urchins day-1, 0.14 ± 0.15 red urchins day-1, 0.34 ± 0.41 mussels day-1, and 
0.01 ± 0.03 cucumbers day-1 (means ± SD; Figure 4.3). When grouping urchin species together, 
Pycnopodia consumed on average 0.91 ± 0.63 urchins day-1.
98
Figure 4.2: Proportional contribution of each prey type to individual Pycnopodia. Each column 
represents a single Pycnopodia, letters under each column indicate original collection location (A
= Eagle Cove, B = O’Neal Island, C = Pt. Caution, D = Goose Island, E = FHL Docks). Numbers
on top of each column indicate the number of total prey items consumed by each Pycnopodia. 
Figure 4.3: Mean number of prey items consumed per day by Pycnopodia across all cafeteria 
experiment trials. Bars indicate standard error and grey circles are individual consumption 
numbers for each of the 11 Pycnopodia tested. 
99
Pycnopodia tended to consume smaller purple and red urchins than the mean size of 
those species that were offered (linear mixed effects, DF = 1, F Ratio = 4.14, p = 0.0432). They 
also consumed slightly larger green urchins than the mean size offered; however, this effect was 
not statistically significant (Figure 4.4). There was no interaction between urchin species and 
size preferences. The mean size of consumed urchins was identical for green (58.9 ± 10.7 mm) 
and purple (58.9 ± 10.1 mm) urchins, the two urchin species most often consumed.
Figure 4.4: Test diameter of three urchin species that were either consumed (purple) or offered 
and not consumed (green) across all cafeteria feeding trials. 
Time-to-first consumption of a prey species, order of species consumed, and total number
of species consumed also varied among Pycnopodia individuals (Figure 2.5). Mean Rodgers’ 
preference calculated as AUC indicated that mussels were the most preferred prey, followed by 
nearly equal preference for green urchins and purple urchins, then red urchins, and lastly, 
100
cucumbers (Table 4.3). This preference was statistically significant  with mussels, purple 
urchins, and green urchins being the most preferred prey (Kruskall-Wallis, chi-square(1, 4) = 
16.65, p = 0.002; Dunn ranking p < 0.05). When considering individual Pycnopodia preferences,
mussels were most often the number one ranked prey item by Rodgers’ preference index (n = 6), 
followed by purple urchins (n = 4), then green urchins (n = 1; Table 4.4).
Figure 4.5: Cumulative consumption of all prey species by Pycnopodia. Boxes are individual 
Pycnopodia, and number of total prey items of each type consumed are listed by trial day (1-10). 
101
Table 4.3: Mean rank of preference and standard deviation for each of the offered prey items 
(chiton is excluded as it was never consumed and had a value of zero). Prefence is calculated as 
area under the curve (AUC) as suggested by Rodgers 1990. 
Table 4.4: Complete AUC values for each Pycnopodia and associated preference ranking. 
102
Shannon diversity values for Pycnopodia diet ranged from 0 (only a single prey item 
consumed) to 1.17 (Figure 4.6). The expected Shannon value if all prey were consumed in the 
proportion in which they were offered (0.2) is 1.61. Measures of evenness ranged from 0 to 0.77 
(Figure 4.6), where an evenness value of 1 would indicate that all prey were consumed in the 
same quantity.
Fig 4.6: Shannon diversity and evenness of Pycnopodia diet in cafeteria feeding experiments. 
Column labels are the same as in Figure 2. Dotted red line in Shannon diversity panel indicates 
expected value if all prey were consumed in the proportion in which they were offered. 
Pycnopodia did not demonstrate consistent prey preferences dependent upon where they 
were collected from (PERMANOVA, PERM = 9999, DF = 4, SS = 0.952, F = 1.117, p = 
0.3598).
103
4. DISCUSSION
This study evaluated the natural prey spectrum as well as prey selection of the sunflower 
sea star Pycnopodia to better asses its role as a generalist predator in intertidal and shallow 
subtidal environments across its North Pacific range, and to explicitly address the assumptions 
and knowledge gaps surrounding Pycnopodia feeding. Their natural prey spectrum in a meta-
analysis of published literature showed bivalves and urchins to be the most common prey, 
although diet was variable. These findings were supported through a cafeteria experiment that 
revealed great variability in prey selection on the level of individual Pycnopodia, both in 
quantity of prey consumed and identity of those prey. Despite the variability, Pycnopodia did 
exhibit specific preference for particular types of prey, particularly mussels and some sea urchin 
species. Combined, these results confirm Pycnopodia to be a generalist predator, but the fact that
some prey were preferentially consumed, especially sea urchins, could have implications for 
Pycnopodia’s role in trophic cascades in kelp forest systems.
Pycnopodia‘s preference for urchins and bivalves in both the meta-analysis and cafeteria 
experiments is particularly interesting in the context of possible effects on ecosystem function, as
these prey fulfill very different roles. Urchins are grazers that can denude kelp forests creating 
alternate stable state urchin barrens (Filbee-Dexter & Scheibling 2016) and bivalves are filter 
feeders important in pelagic-benthic coupling and—dependent upon bivalve species—can also 
be a foundation species (i.e., mussels; Seed 1969, Newell 2004). The ecological role of these 
prey can also shift across latitude and seascape conditions. For example, although purple urchins 
have been directly implicated in kelp forest removal across California (Bonaviri et al. 2017, 
Rogers-Bennett & Catton 2019, Eisaguirre et al. 2020), other species such as red urchins and 
green urchins are identified as the primary kelp grazers at higher latitudes (Schultz et al. 2016, 
104
Burt et al. 2018). Likewise, although red urchins are the dominant urchin species in the San Juan 
Archipelago of Washington State, they are not observed to create urchin barrens as they are able 
to subsist on drift algae provided by strong tidal currents (Lowe et al. 2015). Considering that the
range of Pycnopodia and urchin species linked to kelp forest loss stretches from Baja to the 
Aleutians (1000’s of km), and across multiple seascape conditions from wave-swept coastal 
environments to tidal-current dominated inland seas, we must explicitly consider the effect of 
scale and how ecological roles shift through space (Byrne et al. 2019).
One of the environmental conditions that varies across Pycnopodia’s geographic range is 
temperature. Temperature gradients can influence the strength of predator control in marine 
communities, and to mediate trophic interactions and behaviors (Byrne et al. 2019, Ashton et al. 
2022). The geographic range of Pycnopodia crosses a temperature gradient from the relatively 
warmer waters of the California Current in the south (~17 °C; Mauzole et al. 2020) to the cooler 
waters of the Alaska Current in the north (~9 °C; Bograd et al. 2005, Hickey & Royer 2010). Our
meta-analysis, while spanning some oceanographic regimes over the 24° of latitude, did not 
allow us to test for the effects of temperature on observed Pycnopodia feeding across their range 
because of low replication and missing spatial coverage. Very little urchin consumption, 
however, was observed at the very southern-most end of the dataset (Pacific Grove, CA). 
Pycnopodia predation on urchins is mediated in-part by a temperature refuge in which urchins 
are able to persist at higher temperatures beyond those tolerated by Pycnopodia (>14 °C; 
Bonaviri et al. 2017). As marine heatwaves increase across parts of Pycnopodia’s range (25 day 
mean duration, + 0.8 °C above 99th local percentiles; Frölicher et al. 2018), we can expect the 
size and frequency of this urchin temperature refuge to increase and the metabolic and behavioral
responses of Pycnopodia to change locally based on exposure to these warmer waters. This could
105
provide additional release from predation for grazing urchins across a wide swath of the North 
American West Coast, adding additional grazing pressure to already temperature-stressed kelp 
forests (Starko et al. 2022). In contrast, warmer temperatures have been shown to increase prey 
consumption by sea stars through increased movement, resulting in more prey encounters, 
shorter handling times, and impaired prey escape responses (Barbeau & Scheibling 1994). This 
could suggest that Pycnopodia consumption of urchins will increase as long as the environmental
temperatures are within Pycnopodia’s thermal tolerance window, and then decrease as they 
approach and exceed that limit. However, it is unknown if this optimum ‘thermal window’ would
persist across large enough areas for sustained periods of time, as marine heatwaves even at the 
most northern end of Pycnopodia’s range begin to reach their thermal tolerance (>14 °C; 
Bonaviri et al. 2017, Bograd et al. 2010). As all of the data in our meta-analysis were collected 
from 1968-1983, the recent doubling in marine heatwave intensity and frequency (1982-2016; 
Frölicher et al. 2018) and any associated feeding preference changes would likely not be 
reflected in those observations. Any conclusions regarding the strength of Pycnopodia predation 
pressure in the future would need to explicitly incorporate the effects of temperature. Therefore, 
exactly where and how Pycnopodia’s specific performance curve intersects with the urchin 
temperature refuge is still unclear and requires further study.
The nutritional value of resources can be an important factor in consumer choices (Cruz-
Rivera & Hay 2000) but needs to be balanced with the cost of acquiring that prey, according to 
the optimal foraging theory (Pyke et al. 1977). The metabolic cost of handling and consuming 
urchins by Pycnopodia in the laboratory has been shown to be greater compared to the handling 
and consumption of clams (McGaw & Twitchit 2011), and this metabolic trade-off could be an 
explanation for Pycnopodia’s preference for mussels in our cafeteria trials. The nutritional 
106
content of mussels (the most preferred prey in our cafeteria experiment) tends to be higher only 
when considering protein content of tissues (Dare & Edwards 1975), but urchins (here: purple 
and green, the second most preferred prey) tend to be more nutrient rich with respect to 
carbohydrates and lipids (gonad tissue; Liyana-Pathirana et al. 2002). Lipids in particular are 
energy-dense macronutrients that provide calories, as well as fatty acids, many of which are 
essential to cellular function (Budge et al. 2006). These essential fatty acids can only be obtained
by consumers through consumption, and thus may provide added incentive in prey choice when 
considering what a predator chooses to eat. It has been shown that selective consumption of prey
—both prey identity, and body parts of prey—based on macronutrient content is a widespread 
phenomenon (Kohl et al. 2015). As mussels were the most preferred prey in our cafeteria trials, it
could suggest that Pycnopodia prefer protein-rich prey items over lipid-rich ones. In addition, it 
has been demonstrated that Pycnopodia is unable to discern between purple sea urchins that have
a high gonad index versus a low gonad index (a relative measure of lipid content; Galloway et al.
2023). Therefore, it is possible that either: protein content rather than lipid content is a stronger 
driver of Pycnopodia feeding preferences or; that trade-offs associated with capturing a lipid-rich
prey (i.e., urchins) is high and other factors including prey defenses and prey escape behaviors 
may be stronger determinants of Pycnopodia choices. Given spatial and temporal variability of 
invertebrate prey associated with kelp forests (e.g., Lamy et al. 2018), prey defenses and escape 
behaviors could be prominently influence Pycnopodia prey choices if a nutritious and less costly 
to handle prey (e.g., bivalves) is less abundant.  
Prey characteristics such as spine length and test size did not seem to affect Pycnopodia 
preferences when considering green and purple urchins, which were preferred nearly identically 
across cafeteria trials despite the longer spines and thicker test of purple urchins. Spine length, 
107
however, likely was a factor when considering red urchins, as red urchins possess much longer, 
thicker spines than purple or green urchins (e.g., Tegner & Dayton 1981), and only the smaller 
red urchin individuals were consumed in cafeteria trials. Other urchin predators selectively feed 
on purple over red urchins, and the shorter spine length in purple urchins was the primary factor 
when selections were made (Tegner & Levin 1983).
In our cafeteria experiment all prey were offered with no opportunities to employ 
predator escape tactics as they were contained in a single enclosure with no refuge provided. Sea 
cucumbers were observed to react strongly to physical contact with Pycnopodia and would often 
quickly move away. This relatively quick escape response could partially explain the low-
consumption rate of sea cucumbers by Pycnopodia in these trials. In the wild, sheltering behavior
by sea urchin species, whether under rocks (Nichols et al. 2015) or under larger conspecifics 
(Nishizaki & Ackerman 2007), are the most-commonly employed escape strategies and they 
likely influence Pycnopodia predation behavior when attempting to feed on them. As sessile 
organisms, mussels use byssal attachment and shell strength as a primary deterrent to predation, 
as well as settling among conspecifics to reduce chances of predation. The first two strategies 
were not enough to provide sufficient protection from Pycnopodia in feeding trials, and the third 
strategy was not possible in the experimental setting. The chiton species offered in our 
experiment was also not fast enough to ‘run away’ from Pycnopodia, which are some of the 
fastest crawling sea stars (Montgomery 2014); however, chitons were likely able to avoid 
predation through their ability to adhere firmly to the bottom of the enclosure with a muscular 
foot. Feeding scars were apparent on several of the chitons over the course of the experiment, 
indicating that the Pycnopodia attempted to consume them, but was unable to remove them from 
108
the substrate or penetrate their leathery protective mantle tissue completely covering the dorsal 
side in this species.
Two additional sources of observed variation in the meta-analysis of Pycnopodia prey 
preferences could be explained by 1) the habituation of Pycnopodia to prey items commonly 
found in their habitat of origin (i.e., urchins in kelp forests, bivalves in sand flats, etc.); or 2) 
abundance and identity of predatory competitors that may shift Pycnopodia preferences to 
particular prey types. It has been suggested that echinoderms are capable of associative 
conditioning and habituation (Freas & Cheng 2022). We did not see differences in prey selection 
based on site of collection in our experiment; however, all but one Pycnopodia came from hard-
bottom substrate, which did not provide the statistical power to differentiate between substrate 
types. It is also important to note that our Pycnopodia were fed a maintenance diet of mussels 
and this could potentially be a source of diet habituation, however, whether Pycnopodia actually 
habituate to diets and the time it would take for them to habituate is unknown and a topic of 
future study. Predatory competitors may also change feeding preferences in Pycnopodia through 
intra- and inter-specific interactions (Gaymer et al. 2002). Sea otters, spiny lobsters, and 
sheephead are all present in the southern end of Pycnopodia’s range, and are known to be purple 
urchin predators (Eisaguirre et al. 2020). Likewise, sea otter populations have been recovering 
across much of their range, which overlaps with Pycnopodia‘s range, though otter distribution is 
patchy (Bodkin 2015). Even if there are competitive interactions that drive prey selection in 
Pycnopodia, those sorts of phenomena are notoriously difficult to quantify and observed prey-
switching may simply be a response to the abundance of prey in a particular habitat (Vallina et 
al. 2014) rather than competitive interactions.
109
As Pycnopodia populations begin to recover after their decline during sea star wasting, 
understanding how their trophic interactions affect foundation species such as kelp is critical to 
conserving these habitats. This is especially true as Pycnopodia are also found in association 
with other foundations species like seagrass (Gravem et al. 2021), and may more generally play 
important roles in maintaining ecosystem health. Based on our study we cannot conclude about 
Pycnopodia feeding preferences based on habitat type (i.e., hard bottom versus soft bottom), 
though this is an important consideration when determining the ecological ‘reach’ of Pycnopodia
in nearshore environments, especially in light of their preference for bivalves. This study 
provides an additional piece of the puzzle by synthesizing our knowledge about Pycnopodia 
feeding preferences in the wild, and quantifying relative preferences for common prey species 
that they would encounter in situ. Further studies building on these results include integrating the
effects of climate change, especially considering the predicted increase in marine heatwaves 
(Frölicher et al. 2018), on Pycnopodia’s feeding behavior, determining their feeding preferences 
across more diverse suites of prey, and parsing out their relative role as a predator and scavenger.
BRIDGE
Chapter IV quantified the feeding preferences of the generalist predator 
Pycnopodia helianthoides. In the next chapter I consider how non-consumptive effects 
of Pycnopodia can change the grazing and foraging behavior of one of their preferred 
prey items, the purple sea urchin Strongylocentrotus purpuratus. 
110
CHAPTER V
THE SUNFLOWER SEA STAR PYCNOPODIA HELIANTHOIDES REDUCES SEA URCHIN
GRAZING BEHAVIOR THROUGH NON-CONSUMPTIVE EFFECTS
The study contains previously unpublished coauthored material and was conceived by 
myself with conceptual support from coauthors Aaron Galloway and Sarah Gravem. I carried out
all field collections with assistance from Aaron Galloway. Lab experiments were conducted by 
myself. Video data extraction was done by coauthor Ethan Porter-Hughes. Statistical analyses 
were developed by myself with assistance from Sarah Gravem. I wrote the manuscript with 
editorial input from all authors.
1. INTRODUCTION
Predator-prey interactions are among the strongest drivers shaping ecosystem processes, 
with often far-reaching consequences for community stability and resilience across spatial scales 
(Paine 1966, Preisser et al. 2005, Kissling & Schleuning 2015). This is particularly true of 
strongly interacting species that affect associated community diversity and production through 
trophic cascades, in which one species indirectly controls another (Ives et al. 2005, Byrnes et al. 
2006). While these interactions are frequently mediated in the form of lethal or consumptive 
density-mediated indirect interactions (DMII), it is also possible for non-lethal phenotypic or 
behavioral trait-mediated indirect interactions (TMII) driven by the predators to influence 
ecological outcomes (Preisser et al. 2005, Harding & Scheibling 2015).
TMII in species relationships often act through the alteration of traits of the prey species 
111
involved, including changes to life history or trophic roles in the ecosystem (Bolker et al. 2003). 
These non-consumptive effects (NCE) then scale up to the ecosystem level as an analog to the 
classic trophic cascade based on consumptive predation pressure (e.g., Ripple et al. 2016). 
Frequently, TMII acts via NCEs through the modification of a grazing prey species’ feeding 
behavior, relieving the grazers' food source from grazing pressure (Haggerty et al. 2018, Gravem
& Morgan 2019). The so-called ‘ecology of fear’ can have dramatic effects on ecosystem 
productivity, changing foraging times and grazing rates of important consumers in a suite of 
habitats (Brown et al. 1999).
In marine ecosystems, NCEs have been implicated in a range of predator-prey 
interactions (Werner & Peacor 2003). Predatory cues propagate through these systems via 
various pathways such as visual, auditory, or chemosensory pathways. Chemosensory cues 
proliferate as water flow carries the ‘cue’ of the predator through the water and can change the 
behavior of organisms across the seafloor as they encounter it (Kats & Dill 1998). As such, a 
single predator may have a ‘larger’ effect (i.e., multiple prey species responding) via this non-
consumptive interaction than compared to consumption of a single prey item. Chemosensory 
cues can act as NCEs and may trigger various physiological and behavioral responses in the 
prey. Physiological responses may include metabolic changes (Hawlena & Schmitz 2010), 
reduced fecundity (Bourdeau et al. 2016), and impaired growth (Werner & Peacor 2003), while 
behavioral changes can include reduced foraging time (Werner & Peacor 2003), increased 
crypsis (Scheibling & Hamm 1991), and more frequent escape response (Freeman 2006). All 
these effects can be experienced individually, or simultaneously, and can have serious 
consequences for the prey species. The frequency of encounters with a predator cue may also 
produce various responses in the potential prey. Continuous versus irregular exposure to a 
112
predator cue could result in a prolonged high-stress state on one hand, or gradual habituation on 
the other (Van Dievel et al. 2016, Dehaudt et al. 2019). In addition, the preexisting physiological 
state of an organism can modulate the specific response to a predator. This is seen, for example, 
when a hungry prey species is willing to take more risk of predation than a fed one to obtain food
(Vadas et al. 1994).
The recent population collapse of the generalist invertebrate mesopredator Pycnopodia 
helianthoides (hereafter Pycnopodia) due to sea star wasting disease has been implicated as a 
major contributing factor to the proliferation of the grazing purple urchin, Strongylocentrotus 
purpuratus, after its release from Pycnopodia predation across much of the West Coast of North 
America (Burt et al. 2018, Harvell et al. 2019, Hamilton et al. 2021, Galloway et al. 2023). This 
effect has been most pronounced along the central and southern end of the range of both 
Pycnopodia and purple urchins along the Oregon and California coasts, although similar patterns
have been observed between Pycnopodia and red urchins (Mesocentrotus franciscanus) and 
green urchins (Strongylocentrotus droebachiensis) farther northward into British Columbia 
(Schultz et al. 2016, Burt et al. 2018, Rogers-Bennett & Catton 2019).
Pycnopodia is a common predator of a variety of sea urchins, including S. purpuratus 
(Mauzey et al. 1968, Herrlinger 1983, Shivji et al. 1983), which has a less effective defense 
strategy against Pycnopodia predation than other urchin species (Moitoza & Phillips 1979). 
NCEs from Pycnopodia chemical cues may add to the response of urchins in addition to removal
via direct consumption (Freeman 2006). Pycnopodia is only one among a suite of potential 
predators for purple urchins, although the geographic extent and identity of urchin predators 
change across their range. The spiny lobster (Panulirus interruptus) and California sheephead 
(Semicossyphus pulcher) are constrained to the southern end of the purple urchin’s range (Tegner
113
& Dayton 1981), while the wolf eel (Anarrhichthys ocellatus; Marliave 1987) and sea otter 
(Enhydra lutris; Smith et al. 2021) are found in low densities or are not uniformly distributed 
across seascape. As such, the coincident population decline of Pycnopodia and concurrent 
increase in urchin populations is thought to be a causative relationship, though additional factors 
including warming water temperatures and increased urchin recruitment are also implicated in 
the recent urchin population surge (Burt et al. 2018, Rogers-Bennett & Catton 2019).  
The decline of kelp forests by up to >90% across the central North American West coast 
is attributed, in part, to a predation release on grazing urchins and has been suggested to be a 
putative trophic cascade (Schultz et al. 2016, Rogers-Bennett & Catton 2019). The resulting 
mosaic of kelp forests and urchin barrens–which are considered an alternate stable state–are 
marked by decreased primary productivity, low food web complexity, and a reduction in 
economically important species such as abalone (Filbee-Dexter & Scheibling 2014, Schultz et al.
2016, Rogers-Bennett & Catton 2019). Urchins within these urchin barrens can maintain this 
alternate stable state when the reduced kelp biomass leads to starvation, and constant grazing 
inhibits the regrowth of kelp and other macroalgae (Estes & Duggins 1995, Filbee-Dexter & 
Scheibling 2014, Spindel et al. 2021, Dolinar & Edwards 2021).  
Although direct consumptive effects of Pycnopodia on urchins is an important pathway 
of grazer control (Bonaviri et al. 2017, Galloway et al. 2023), behavioral responses by urchins to 
predation cues suggest that NCEs may also play a key role in suppressing sea urchin grazing 
behavior (Moitoza & Phillips 1979, Bernstein et al. 1981, Duggins 1983, Lee et al. 2016). In 
addition, higher concentrations of urchins in urchin barrens can increase their hunger-state and 
their response to food and predator cues, resulting in potentially complex behaviors (Parnell et al.
2017). Yet, the effect of NCEs between Pycnopodia and sea urchin behavior and grazing rates 
114
has not been explicitly quantified. Understanding how and to what degree Pycnopodia may alter 
sea urchin grazing can provide important insights into their role in maintaining healthy kelp 
forest ecosystems through TMII. To address these knowledge gaps, we here asked the questions: 
1) Do NCEs between Pycnopodia and S. purpuratus alter the feeding rates of the urchins 
(feeding experiment)?, and 2) Does the hunger-state of S. purpuratus change their response to 
food in the presence of a predator cue (movement experiment)?
2. METHODS
2.1 Feeding experiment organism collections and care
Purple sea urchins (S. purpuratus) were collected at Lopez Island (48.4558, -122.9383) in
the San Juan Archipelago of Washington State on September 11, 2020 (71 urchins) through a 
permitting arrangement between the Friday Harbor Labs and Washington State (Washington 
State House Bill 68, R.C.W.28.77.230, 1969 Revision R.C.W.28B.20.320). Collections were 
made by hand on SCUBA between 3-8 m below mean lower low water (MLLW) in bull kelp 
(Nereocystis luetkeana) stands on bedrock. Urchins were then held for 2 days in flow-through 
sea water tanks (60 cm x 90 cm x 30 cm, WHD) prior to experimentation beginning on 
September 13, 2020. Urchins were shaded from direct sunlight and fed N. luetkeana blade tissue 
ad libitum while in holding.
Pycnopodia were collected at Goose Island in the San Juan Archipelago (48.4575°, -
122.9544°) on July 29, 2020 at a depth of 6 m below MLLW through the same permitting as 
above. Two Pycnopodia were collected to be used in all feeding experiments because of the 
overall low abundance of this species after the sea star wasting decline and to minimize possible 
local, ecological effects of removing an endangered species. Pycnopodia were transported 
115
following lab and handling protocols described in Hodin et al. (2021) in covered 5-gallon 
buckets filled with seawater and placed in shaded flow-through seawater tanks for 6 weeks (120 
cm × 40 cm × 45 cm WHD) prior to experimentation. They were fed a maintenance diet of 2 
mussels (Mytilus spp.) every other day prior to experimentation, and in-between trials during 
experimentation following lab and handling protocols described in Hodin et al. (2021).
2.2 Feeding experiment setup
To determine if there are non-consumptive effects of Pycnopodia on sea urchin feeding 
rates, five replicate feeding trials were run between September 13 – October 2, 2020, each lasting
from 51-69 h using a total of two Pycnopodia to generate chemical cues and 71 S. purpuratus. 
We used only two individual Pycnopodia across multiple trials for the reasons of minimizing 
ecological impacts listed above. All urchins were used in the experiment only once to ensure 
independence of replicates. Sea urchin test diameters were measured with hand calipers to the 
nearest millimeter, and urchins were assigned to two experimental groups (Pycnopodia exposure 
or control). Experimental setup, including organism arrangement, tanks, and plumbing, were 
modified after the initial Trial 1 to reduce stress on the Pycnopodia, but the effective predator 
cue treatments remained consistent between trials, so they were analyzed as replicate 
experiments. For Trial 1, five urchins were placed in each of three tanks (circular: 50 cm depth x 
104 cm diameter); one of these tanks contained a caged Pycnopodia so that chemical cues were 
released into the tank but no physical interaction with the urchins was possible. The other two 
tanks served as controls without a Pycnopodia. For Trails 2-5, seven urchins were placed into 
each of two tanks (circular: 30 cm depth x 104 cm diameter). One of these tanks received a 
predator cue via effluent from a header tank containing Pycnopodia (Figure 1A) to the urchin 
116
tank while the other served as a control and received plain seawater (no cue). Urchins within a 
tank in all treatments were individually contained within buckets perforated with multiple small 
holes on the sides and top of each container (<1 cm diameter each) to maintain water flow within
each, while preventing interactions among urchins in the tank (Figure 1A). Water flow rate into 
experimental tanks was 3.89  ± 0.46 liters/minute, water temperature during the experiment was 
10.0  ± 0.16 °C, salinity 30.54 ± 0.81, as measured by the Friday Harbor Labs Ocean 
Observatory, and ambient light was an 11:13 h light:dark cycle.
Each urchin in the individual tank buckets was fed 12 standardized circular pieces of kelp
'confetti' (21 mm diameter) created by using a cork punch on vegetative Nereocystis luetkeana 
(hereafter kelp) blade tissue. A subsample of kelp confetti was weighed at the beginning and end 
of the experiment to estimate biomass of kelp consumed, and a subset of trials (n = 4) contained 
additional control confetti held separately from urchins that were weighed before and after each 
trial to detect changes in confetti weights that were not related to urchin feeding. Urchins in each 
experimental unit were fed kelp confetti ad libitum and their consumption rates tracked by 
counting number of intact confetti pieces and estimating proportion of partially-consumed pieces
present at three time points per day (08:00, 14:00, 20:00) for up to three days. Kelp confetti had a
mean biomass of 0.341 ± 0.030 g per disc (n = 49), and control confetti weights were not 
statistically different before and after trials (linear mixed-effects model [lme], df = 92, t = 0.45, p
= 0.65) and so were not used as a correction factor when considering biomass loss. Any 
consumed confetti were replaced with fresh confetti maintaining a total of ~12 pieces per urchin. 
Trial 1 had to be stopped after 51 hours, while all remaining trials were run for 69 hours.
117
Figure 5.1: Setup for urchin feeding (A) and urchin movement (B) experiments. All treatment 
combinations are shown for each experiment in tables. The urchin grazing behavior setup depicts
the ‘fed urchin with algae’ treatment, and the flow diagram shows general flow created by water 
inflow located at the center-bottom of the arena, and outflow at the center-top. Two identical 
‘urchin arenas’ were used for the grazing behavior experiment to allow for tank cleaning 
between trials, but only one is shown here for clarity.
2.3 Movement experiment organism collections and care
S. purpuratus were collected at Lopez Island (48.4558°, -122.9383°) in the San Juan
Archipelago of Washington State on June 11, 2021 (49 urchins) under FHL permitting (see 
above). Collections were made by hand using the same protocols as above, and urchins then held
118
in flow-through sea water tanks (circular: 30 cm depth x 104 cm diameter) before being moved 
into the experimental tanks (rectangular: 60 cm x 90 cm x 30 cm WHD).
A single Pycnopodia was collected under the FHL docks (48.54532, -123.01192) on July 
3, 2021 via SCUBA at a depth of 4 m below MLLW under FHL permitting. Only a single 
Pycnopodia was used as a source of predator-cue for this experiment to minimize effects of 
collection on wild populations of this endangered species. The sea star was transported in a 
shaded 5-gallon bucket filled with seawater and then held in a partially shaded flow-through 
seawater tank (120 cm × 40 cm × 45 cm WHD) prior to experimentation.
2.4 Movement experiment setup
S. purpuratus (n = 49) were measured for diameter using calipers and separated
haphazardly into two groups: ‘fed’ and ‘starved'. ‘Fed’ urchins were fed kelp, ad libitum, and 
starved urchins were not fed for seven weeks while in holding. This time frame has been shown 
to be sufficient to result in different gonadal conditions among urchin feeding groups (Galloway 
et al. 2023). At the end of the starvation period all urchins were used in individual trials, placing 
starved (n = 25) and fed urchins (n = 24) separately in one of two identical circular aquarium 
‘arenas’ (30 cm depth x 104 cm diameter), respectively, in which individual urchin movement 
and behavior was tracked over one hour. There were a total of eight treatments in a factorial 
design crossing urchin treatment (fed vs. starved), Pycnopodia cue (cue vs. no cue), and kelp 
treatment (kelp vs. kelp replica; Figure 1B). The Pycnopodia cue was delivered as effluent from 
a header tank containing one Pycnopodia (120 cm × 40 cm × 45 cm WHD) and ‘control’ no-cue 
water from a header tank with no Pycnopodia (same dimensions). In kelp-present trials, a 55 cm 
x 4 cm kelp strip was wrapped around an inverted perforated sampling cup through which the 
119
header source water flowed into the experimental arena. In no-kelp control treatments a kelp 
replica (55 cm x 4 cm plastic sheeting material) was wrapped around the entry point of the 
header water source. In all cases, the centrally placed header Pycnopodia cue inflow and 
kelp/replica location is referred to as the ‘cue source’.
All inflow of water during each 1-h trial came from the header tank. Water temperature 
(12.9 ± 1.19 °C), and light flux (1922 ± 1166 lux) were measured using a HOBO TidbiT v2 Data
Logger (Onset, Bourne, MA, USA), and salinity (31.1 ± 1.46) was measured using a hand 
refractometer. Experimental flow rates were standardized to 2.5 L/min. Patterns of flow diffusion
within each experimental unit were visualized intermittently between trials using fluorescein dye 
injected at the outflow of the header tank to confirm constant flow and even distribution of the 
cue. Each treatment was run at least six times with naive urchins for all eight possible 
combinations, and one trial was rerun due to a malfunctioning camera, for a total of 49 individual
trials. Urchins were observed in 1-min intervals for each 1-h trial for the following target 
behaviors: movement [not moving or moving]; interacting with central cue [foraging, not 
foraging; foraging was defined as at least one tube foot touching the cue source (kelp or kelp 
replica)]. Quantity of kelp consumed (in trials with kelp) was calculated as reduction in surface 
area of experimental kelp strip from known dimensions after the trial was concluded using 
photos taken with a scale bar and processed in ImageJ (Schneider et al. 2012). Experimental 
arenas were drained, rinsed with fresh water, and scrubbed by hand between all trials to 
eliminate cross contamination of chemosensory cues.
Trials were also recorded with GoPro cameras (HERO4 Silver; San Mateo, CA, USA) 
with a medium FOV setting. Videos of each trial were processed in the video analysis software 
Tracker (Brown et al. 2021) using the AutoTracking function. The path of each urchin was 
120
extracted as x and y coordinates in 3.3-sec increments and calibrated to a bullseye of known 
dimensions underneath the urchin. The bullseye was correlated with a center of mass function in 
Tracker to maximize accuracy. Total distance traveled was calculated for each urchin using 
adehabitatLT (Calenge 2006).
2.5 Statistical analysis
Data were summarized, quality controlled, and visualized using R (R Core Team 2022) in
RStudio (RStudio Team 2022) with the tidyverse, adehabitatLT, and viridis packages (Calenge 
2006, Wickham et al. 2019, Garnier et al. 2021). Statistical analyses were performed in R with 
the lme4 package (Bates et al. 2015) and in JMP v17.0.0 (SAS Institute Inc., Cary, NC, USA).
2.5.1 Feeding experiments – Differences in mean amount of kelp consumed per urchin per hour 
between treatments over the course of each trial were tested with a linear mixed-effects model 
(lme). The response variable of mean kelp consumed per hour was log+1 transformed to address 
variance heteroscedasticity. Fixed effects included: treatment, hour since beginning of the trial, 
and their interactions. Random effects included: urchin identity nested within treatment and trial 
tank, and trial tank nested within treatment, to account for any variation not attributed to the 
fixed effects.
2.5.2 Movement experiment – Urchin movement behaviors were averaged over 5-min increments
to detect any changes in behavior over the course of each individual trial. Behaviors were 
assigned binary classifications for each urchin, with 0 = not moving or not foraging, and 1 = 
moving or foraging to give proportion of time performing each behavior within each increment. 
121
Both proportion of time moving, and proportion of time foraging were tested with lme. Fixed 
effects for both responses included: urchin status (fed, starved), Pycnopodia cue (present, 
absent), kelp cue (present, absent), time increment (5-60 in 5-min steps), and the full interaction 
of all these fixed effects. Random effects included trial arena nested within urchin status, 
Pycnopodia cue, and kelp cue, to account for any variation not attributed to the fixed effects. 
Feeding on kelp, when it occurred, was tested with a linear model.
Total distance traveled by urchins during each trial were tested with lme. Fixed and 
random effects were the same as above; however, time increment was not included as the 
resolution of the data was only on the level of a per-trial total.
3. RESULTS
3.1 Feeding experiment
 Urchins exposed to a waterborne Pycnopodia cue consumed 53% less kelp per hour on 
average (0.127 ± 0.104 g/h) than urchins not exposed to a Pycnopodia cue (0.067 ± .047g/h) 
across all trials (lme, DFden = 8.64, t = 2.64, p = 0.0279; Figure 5.2A, Table 5.1). There was a 
significant interaction between cue treatments and hours since the beginning of the trial (lme, 
DFDen = 538.9, t Ratio = -4.96, p < 0.001), reflecting differing changes in consumption rates 
from the beginning to the end of each trial based on treatment; however, consumption was highly
variable across all treatments. Generally, urchins exposed to a Pycnopodia cue increased their 
rate of consumption over time (3 h: 0.039 ± 0.075 g/h; 69 h: 0.084 ± 0.061 g/h) while urchins not
exposed to a Pycnopodia cue decreased their rate of consumption (3 h: 0.191 ± 0.197 g/h; 69 h: 
0.111 ± 0.105 g/h). Consumption rates became indistinguishable after 27 h (Figure 5.2B-C). 
122
Mean hourly consumption of kelp at 27 h was 0.055 ± 0.089 g/h (Pycnopodia cue) and 0.128 ± 
0.112 g/h (no Pycnopodia cue; lme, DF = 69, t = -3.017, p = 0.00357), and at 33 hr was 0.095 ± 
0.105 g/h (Pycnopodia cue) and 0.139 ± 0.129 g/h (no Pycnopodia cue; lme, DF = 69, t = -1.574,
p = 0.12).  
Figure 5.2: Hourly consumption of kelp by urchins that were exposed (light orange) and not 
exposed (dark purple) to a Pycnopodia cue across the 69 hour experimental period (A). 
Regression lines show the best fit and 95% confidence interval for each group. Differences in 
hourly feeding rate (dotted red box) are shown at 27 hours (B) and 33 hours (C), the time at 
which feeding rates become indistinguishable between the two treatments.
123
Table 5.1: Mixed-effects model of the effect of a Pycnopodia cue on urchin feeding.
3.2 Movement experiment
Across all treatments, starved urchins spent an average of 59.8 ± 49.0% of the experiment
moving within the arena, while fed urchins spent an average of 86.7 ± 34.0% of the experiment 
moving (Figure 5.3A). Urchin status (fed, starved) was the primary single fixed factor to drive 
differences in time spent moving across all treatments (lme, DFDen = 41, t Ratio = 4.83, p < 
0.0001; Table 5.2). Various interactions between urchin status and other effects were also 
statistically important, notably urchin status and algal treatment (lme, DFDen = 41, t Ratio = -
2.20, p= 0.0334), and urchin status and time (lme, DFDen = 531, t Ratio = 3.14, p = 0.0018). 
Pycnopodia cue was only significant in an interaction with urchin status and time (lme, DFDen =
531, t Ratio = 2.05, p = 0.0407).  
Time spent foraging varied widely across treatments with starved urchins spending an 
average of 32.1 ± 46.7% of their time foraging, while fed urchins spent only 9.2 ± 29.0% of their
time foraging (Figure 5.3B). Starved urchins in the kelp-present treatment spent much more time 
Sheet1
Effect of Pycnopodia Cue on Urchin Feeding Rates
Predictor Estimate Std Error DFDen t Ratio Prob>|t|
(Intercept) 0.0342 0.0046 13.6100 7.4500 <.0001
Pycnopodia Cue 0.0108 0.0041 8.6440 2.6400 0.0279
Hour of Experiment 0.0001 0.0001 538.9000 1.4500 0.1465
Pycnopodia Cue * Hour of Experiment -0.0003 0.0001 538.9000 -4.9600 <.0001
Random Effect Var Component Std Error 95% Lower 95% Upper Wald p-Value
Urchin ID / Pycnopodia Cue / TrialTank 0.0008 0.0002 0.0005 0.0011 <.0001
TrialTank / Pycnopodia Cue 0.0000 0.0001 -0.0001 0.0002 0.6362
Residual 0.0009 0.0001 0.0008 0.0011
Total 0.0018 0.0002 0.0015 0.0022
Page 1 124
foraging than other groups, regardless of Pycnopodia cue (56.7 ± 49.6). Single fixed effects 
driving observed patterns of foraging included urchin status (lme, DFDen = 41, t Ratio = -2.89, p
= 0.0062), algal treatment (lme, DFDen = 41, t Ratio = -3.67, p = 0.0007), and time (lme, DFDen
= 531, t Ratio = -2.74, p = 0.0063; Table 2). Various interactions were observed between effects, 
though Pycnopodia cue was not included in any of those interactions.
Figure 5.3: Proportional time spent by starved (light teal) and fed (dark orange) urchins across all
treatment combinations in the movement experiment for the proportional time moving (A) and 
proportional time foraging (B). Large circles are overall means per group per treatment, and 
transparent small circles are 5-minute increment averages for individual urchins (12 points per 
urchin = 60-min trial). Vertical dashed line indicates mean proportional time spent doing each 
behavior for all urchins in the behavioral control treatment (no Pycno, no Kelp). Total distance 
traveled (C) in each of the four treatments by starved (light teal) and fed (dark orange) urchins. 
Large circles indicated group mean for urchin conditions (starved, fed), and smaller points 
indicated values for individual urchins within each treatment. The total quantity of kelp 
consumed in kelp treatments (D) were not different between Pycnopodia cue treatments between
starved urchins. Fed urchins never consumed kelp in any treatment.
125
Table 5.2: Mixed-effects model of the effect of a Pycnopodia cue, urchin status (starved, fed), 
and algal treatment (kelp offered, not offered) on the movement, foraging, and distance traveled 
of a sea urchin.
Sheet1
Effect of Pycnopodia Cue on Urchin Movement
Predictor Estimate Std Error DFDen t Ratio Prob>|t|
(Intercept) 0.7391 0.0330 86.94 22.37 <.0001
Urchin Status 0.1315 0.0273 41 4.83 <.0001
Pycnopodia Cue -0.0114 0.0273 41 -0.42 0.6778
Urchin Status * Pycnopodia Cue 0.0184 0.0273 41 0.67 0.5046
Algal Treatment 0.0420 0.0273 41 1.54 0.1314
Urchin Status * Algal Treatment -0.0600 0.0273 41 -2.20 0.0334
Pycnopodia Cue * Algal Treatment 0.0190 0.0273 41 0.70 0.4887
Urchin Status * Pycnopodia Cue * Algal Treatment 0.0115 0.0273 41 0.42 0.6751
Time -0.0001 0.0006 531 -0.21 0.8302
Urchin Status * Time 0.0018 0.0006 531 3.14 0.0018
Pycnopodia Cue * Time 0.0002 0.0006 531 0.36 0.716
Urchin Status * Pycnopodia Cue * Time 0.0012 0.0006 531 2.05 0.0407
Algal Treatment * Time 0.0010 0.0006 531 1.67 0.0963
Urchin Status * Algal Treatment * Time -0.0018 0.0006 531 -3.15 0.0017
Pycnopodia Cue * Algal Treatment * Time 0.0005 0.0006 531 0.88 0.3768
Urchin Status * Pycnopodia Cue * Algal Treatment * Time -0.0006 0.0006 531 -1.13 0.2598
Random Effect Var Component Std Error 95% Lower 95% Upper Wald p-Value
TrialTank / Urchin Status / Pycnopodia Cue / Algal Treatment 0.0315 0.0080 0.0158 0.0473 <.0001
Residual 0.0576 0.0035 0.0512 0.0652
Total 0.0891 0.0087 0.0743 0.1088
T Effect of Pycnopodia Cue on Urchin ForagingPredictor Estimate Std Error DFDen t Ratio Prob>|t|(Intercept) 0.2474 0.0415 57.67 5.96 <.0001Urchin Status h-0.1099 0.0381 41 -2.89 0.0062Pycnopodia Cue -0.0144 0.0381 41 -0.38 0.7065Urchin Status * Pycnopodia Cue 0.0096 0.0381 41 0.25 0.8027Algal Treatment -0.1397 0.0381 41 -3.67 0.0007Urchin Status * Algal Treatment 0.1182 0.0381 41 3.11 0.0034Pycnopodia Cue * Algal Treatment 0.0394 0.0381 41 1.04 0.3063Urchin Status * Pycnopodia Cue * Algal Treatment 0.0196 0.0381 41 0.51 0.6095Time -0.0014 0.0005 531 -2.74 0.0063Urchin Status * Time -0.0021 0.0005 531 -4.21 <.0001Pycnopodia Cue * Time 0.0002 0.0005 531 0.36 0.7188Urchin Status * Pycnopodia Cue * Time -0.0006 0.0005 531 -1.15 0.2495Algal Treatment * Time -0.0015 0.0005 531 -2.93 0.0035Urchin Status * Algal Treatment * Time 0.0020 0.0005 531 4.02 <.0001Pycnopodia Cue * Algal Treatment * Time -0.0003 0.0005 531 -0.68 0.4956Urchin Status * Pycnopodia Cue * Algal Treatment * Time -0.0007 0.0005 531 -1.44 0.1495Random Effect Var Component Std Error 95% Lower 95% Upper Wald p-ValueTrialArena / Urchin Status / Pycnopodia Cue / Algal Treatment 0.0671e0.0156 0.0364 0.0977 <.0001Residual 0.0450 0.0028 0.0400 0.0509Total 0.1120 0.0158 0.0865 0.1509  total distance traveled by urchins was different between starved (576 ± 225 cm) and 
Effect of Pycnopodia Cue on Urchin Distance Traveled
Predictor Estimate Std Error DFDen t Ratio Prob>|t|
(Intercept) 60.1777 2.8242 4.355 21.31 <.0001
Urchin Status 10.9064 2.8242 4.355 3.86 0.0154
Pycnopodia Cue -3.0424 2.8242 4.355 -1.08 0.3374
Urchin Status * Pycnopodia Cue 5.7438 2.8242 4.355 2.03 0.1060
Algal Treatment 2.5141 2.8242 4.355 0.89 0.4198
Urchin Status * Algal Treatment 3.9976 2.8242 4.355 1.42 0.2243
Pycnopodia Cue * Algal Treatment -3.6239 2.8242 4.355 -1.28 0.2635
Urchin Status * Pycnopodia Cue * Algal Treatment -1.3749 2.8242 4.355 -0.49 0.6499
Random Effect Var Component Std Error 95% Lower 95% Upper Wald p-Value
TrialArena / Urchin Status / Pycnopodia Cue / Algal Treatment 22.7157 71.0509 -116.5414 161.9728 0.7492
Residual 264.5786 65.3478 171.9103 459.3453
Total 287.2943 77.3684 180.3669 528.2971
Page 1
fed (842 ± 203) urchins (lme, DFDen = 4.355, t Ratio = 3.86, p = 0.0154; Figure 5.3C, Table 
126
5.2), and no other fixed effect other than urchin state contributed to those differences.  
Only starved urchins were ever observed to consume the kelp in trials that contained the 
cue, and though they tended to consume more when the Pycnopodia cue was present (mean = 
0.74 ± 0.03 g) compared to when it was not (mean = 0.46 ± 0.39 g), this difference was not 
statistically significant (Figure 5.3D).
4. DISCUSSION
We found that waterborne cues of the sunflower sea star Pycnopodia decreased short-
term feeding rates (~27 h) of the purple sea urchin S. purpuratus, as well as changed urchin 
movement patterns, through non-consumptive effects (NCEs). In one-hour behavior trials, urchin
feeding behavior was further modified by the starvation state of the urchins, with starved urchins 
moving less than fed urchins, primarily to feed on kelp regardless of predator cue presence or 
absence. Urchins in a metabolic state similar to those in barrens, which had been starved for 
seven weeks, would feed on kelp, regardless of a Pycnopodia cue. This suggests that starved 
urchins in urchin barrens may continue to feed on kelp even as Pycnopodia populations begin to 
rebound, which may inhibit the role of NCEs on the recovery of kelp forest habitats from an 
urchin barren state. The current proliferation of purple urchins across the West Coast has 
coincided with kelp declines across much of the same area and direct consumption of kelp by 
urchins, suppression of kelp juveniles, and stressors directly associated with climate change 
including warming waters have also been implicated in this decline (Pfister et al. 2018, Rogers-
Bennett & Catton 2019, Beas-Luna et al. 2020, McPherson et al. 2021). The amelioration of a 
single-origin stressor in the form of over-grazing by urchins, though only one of several multiple 
stressors, could lower the tipping point for alternate stable states of former kelp forests.
127
Pycnopodia is an important member of a diverse range of benthic marine ecosystems, and
has been identified as a key mesopredator in kelp forest communities (Schultz et al. 2016, Burt et
al. 2018). As Pycnopodia begin to recover from sea star wasting disease, they may mitigate the 
impact of urchins that are overpopulated in places where kelp has been extirpated (Galloway et 
al. 2023). Our study showed that there is a measurable and significant non-consumptive effect of 
Pycnopodia on sea urchin feeding rates and behavior that may benefit kelp, particularly on short 
time-scales; these NCEs may have important consequences for kelp ecosystems. As Pycnopodia 
population densities begin to rebound, a larger proportion of grazing sea urchins in subtidal 
communities will be subject to both direct consumptive and non-consumptive interactions, 
potentially reducing grazing rates on a population-wide and a per-capita basis. Although the 
single mass of the average Nereocystis individual is ~1200 g (Stekoll et al. 2006), well beyond 
the feeding capacity of even a dense aggregation of urchins, it has been found that even small 
biomass removal, particularly from the stipe of canopy kelps can reduce kelp resistance to waves
and currents, detaching the entire kelp plant and, thus, removing it from the system (Duggins et 
al. 2001). The release of urchins from the direct effects of Pycnopodia predation is identified as 
one causal link leading to kelp forest decline, as Pycnopodia may consume up to 0.68 ± 0.33 
urchins per day (Galloway et al. 2023). Our work demonstrates that, in addition to this direct 
consumptive effect, the waterborne cues of nearby Pycnopodia likely exert a non-consumptive 
effect on urchins by suppressing grazing rates over 50%. Even if Pycnopodia do not directly 
consume purple urchins at such high rates, as what has been found in the lab, the non-
consumptive effects documented here represents an important reduction in urchin feeding and a 
potentially important TMII, in which Pycnopodia benefit kelp by suppressing urchin grazing.    
128
This work also broadens our knowledge about urchin response to NCEs and how direct 
chemosensory detection of a sea star predator is sufficient to alter the urchins’ behavior. Much 
work to date on the sea urchin response to predation cues has focused on their reaction to 
damaged conspecifics as an alarm cue, rather than the presence of a predator itself (Campbell et 
al. 2001, Spyksma et al. 2017, Belleza et al. 2021). While this is a good proxy for actively 
feeding predators that may break or otherwise damage urchins such as sea otters and spiny 
lobsters, it does not account for the potential effects of predators that are not actively feeding. 
When the effect of a predator cue has been tested directly on urchins, it has been done with 
gastropod (Hexaplex trunculus) and arthropod (Panulirus interruptus) predators (Pagès et al. 
2021, Knight et al. 2022). Similar to Knight et al. (2022), we found that starved urchins tended to
consume kelp whether a predator was present or not, though their study did detect a difference in
consumption between predator treatments in the case of well-fed urchins, while in our movement
experiment fed urchins did not consume kelp in any treatment, including those without a 
Pycnopodia cue. This difference could be a result of a number of factors, including differential 
response to particular predators (Pycnopodia versus lobster), an effect of urchin density cues in 
treatments (tested individually versus with conspecific cue), or a statistical effect (n = 8 in 
Knight et al. (2022) versus n = 71 in this study). In other studies, urchins exposed to gastropod 
predator cues have been shown to exhibit a more linear and fast movement escape response than 
urchin movement without the cue (Pagès et al. 2021). This is in contrast to our study, where we 
did not detect a statistical difference in total distance traveled or proportion of time spent moving
in response to a Pycnopodia cue between fed and starved urchins, though starved urchins moved 
slightly less during the movement experiment compared to fed ones, and there was a slight 
reduction in distance traveled when fed urchins were exposed to a Pycnopodia cue only. This 
129
raises the possibility that response of urchins to a predator cue is highly variable or dependent 
upon other factors such as conspecific density, urchin size variation, or shelter availability 
(Nishizaki & Ackerman 2007, Green 2012).
 Predator speed and hunting style may also affect the response of urchins to a particular 
predator cue. Starved urchins moved less than fed urchins, regardless of Pycnopodia presence, 
though this difference was most pronounced when there was no predator cue but with kelp 
offered. This is most likely a hunger-response by the starved urchins to the kelp cue, which 
suggests that the non-consumptive waterborne Pycnopodia cue was not sufficient to inhibit 
urchin feeding. This may also be a metabolic consequence of the starved urchins as they are 
known to divert energy from reproduction into maintenance and need access to food to quickly 
restore metabolic activity and replenish somatic reserves (Smith & Garcia 2021). Both the 
hunger and metabolic consequences of starvation could indicate that food access despite the 
danger of predation is a stronger incentive than predator avoidance for starved urchins.
The fact that starved urchins increased total distance traveled in the case of a Pycnopodia 
cue compared to no predator cue could suggests predator evasion, although the overall patterns 
we observed were not strong enough to for unequivocal conclusions. The same is not true of the 
fed urchins, which, while moving greater distances than starved urchins in general, had 
decreased movement distances in the presence of a Pycnopodia cue alone. This seems to be 
contrary to a predator flight response evident in starved urchins, and these differences highlight 
the need to further study indirect effects of sea star predators on urchins.
Pycnopodia are generalist scavengers and opportunistic predators that are highly mobile 
within their environment, and are known to elicit escape responses by green and red sea urchins 
(Paul & Feder 1975, Freeman 2006, Montgomery 2014). This mobility may increase the reach of
130
the Pycnopodia’s non-consumptive effects, especially where urchins are congregated in dense 
aggregations. Such aggregations are most common in urchin barrens (Rowley 1989), which 
typically represents urchins in starved conditions (Smith & Garcia 2021, Dolinar & Edwards 
2021). Considering the difference in urchin behavior in terms of feeding and movement in the 
presence of a Pycnopodia cue (see above), the effect of a predator cue would likely be different 
in urchin barrens where the urchins are in a relatively starved state compared to well-fed urchins 
in kelp habitats.  
Studies that have targeted the non-consumptive interactions between Pycnopodia and sea 
urchins have found that urchin size (Freeman 2006), water temperature (Bonaviri et al. 2017), 
and urchin species identity (Duggins 1983) can result in different feeding or escape behaviors. 
Additionally, it has been found that juvenile and small urchins seek shelter by moving closer to 
the spine canopy of larger conspecifics (Nishizaki & Ackerman 2007). This behavior may allow 
urchins that have not reached a size-refuge from Pycnopodia predation to continue feeding, as 
has been seen in cryptic urchins exposed to predator cues (Green 2012). Interestingly, we did not
detect a statistical effect of urchin size across any of our experiments, however, we did detect a 
trend in urchin size and feeding rate in the feeding experiment (Figure 5.4). As we targeted a 
homogeneous diameter size specifically, it’s possible that the size effect was dampened, and a 
greater effect of size would be seen if a larger range of urchins sizes were incorporated into the 
experiment.
131
Figure 5.4: Relationship between urchin test diameter and mean quantity of kelp consumed per 
hour when exposed (light orange) and not exposed (dark purple) to a Pycnopodia cue. The trend 
is not statistically significant, but indicates that size may play a role in urchin feeding behaviors 
when exposed to a predator cue.
This study demonstrated a non-consumptive behavioral effect of Pycnopodia on S. 
purpuratus in laboratory conditions, but the potential ecological impact of this effect would be 
dependent upon in situ conditions and the ability of urchins to detect Pycnopodia cues in nature, 
which are subject to diffusion, wave action, and currents (Chivers et al. 2013). In general, 
however, the importance of water-borne chemical cues mediating ecological responses in the 
marine environment have been well documented (e.g., Zimmer & Butman 2000, Hadfield & Paul
2001). On the one hand, this ‘dilution’ of the predator cue may reduce or negate any indirect 
132
effect that itinerant Pycnopodia may have on urchins. On the other hand, if the detection 
threshold is low enough, it may instead spread the cue over a wide area, affecting a large number
of urchins, especially densely aggregated urchins in a barren. In addition, the frequency and 
duration of these cues may produce different responses in urchins depending on whether they 
become accustomed to a predator cue over either long-term, or repeated short-term exposure. In 
contrast, the accumulated exposure to predator cues has also been shown to “train” marine 
invertebrates in improving their escape response to predators (Rochette et al. 1998). It remains to
be seen which long-term effects Pycnopodia may have on urchins, whether that be persistent 
cryptic behavior (Spyksma et al. 2017), habituation (found in holothuroids; Hamel et al. 2021), 
or something in between. Although we found short-term (3 h to two days) behavioral changes, 
the effects of concentration and periodicity of predator cues require more study. Future directions
to expand our knowledge of how Pycnopodia might indirectly affect sea urchins include: 1) 
What is the strength of a Pycnopodia cue in situ (i.e., at what distance can an urchin detect a 
Pycnopodia in the water)?; 2) What is the starvation threshold at which a sea urchin will risk 
predation to acquire food?; 3) How long will a Pycnopodia suppress the appetite of a sated sea 
urchin (i.e., will a sea urchin acclimate to the Pycnopodia cue over time)?; 4) Does a sea urchin’s
size modulate the risks it will take to acquire food?; and 5) Do multi-species predator cues elicit 
a stronger behavioral response from sea urchins?
Scaling up the effects of individual urchins in our experiments, it is possible that the 
laboratory-observed ~50% reduction in feeding with Pycnopodia waterborne cues could have a 
large effect on kelp persistence. Given that N. luetkeana can grow up to 6% in body size a day 
(Kain 1987), the reduction in urchin grazing intensity could result in a considerable increase in 
biomass build-up of kelp in the presence of Pycnopodia and assist in the long-term re-
133
establishment of local kelp forests (Galloway et al. 2023). Once a kelp forest establishes, the 
delivery of drift kelp through wave and current action can provide a large proportion of urchin 
diet in multiple systems, inhibiting feeding on live kelp plants (Lowe et al. 2015, Kriegisch et al. 
2019). This work provides key findings on the effects of non-consumptive effects between 
Pycnopodia and their S. purpuratus prey, and contributes new knowledge to our understanding 
of the indirect effects of Pycnopodia on kelp. The results show that re-establishment of 
Pycnopodia populations could assist in the re-establishment of currently degraded kelp forests in 
the Northeast Pacific based on indirect predator-prey interactions.
134
CHAPTER VI
CONCLUSION
This dissertation highlights the community ecology of macroalgal forests with a broad 
geographic, taxonomic, and methodological scope, from the scale of algal tissues, to the scale of
regional marine protected areas.  I have identified the unique fatty acid profiles and stable 
isotope content of 31 Antarctic macroalgae, provided a summary of the effects of conservation 
areas on algal assemblages, and provided evidence for non-consumptive effects and feeding 
preferences of the sunflower seastar on grazing sea urchins.  
Chapter II described biomarker profiles from common Antarctic macroalgae across the 
WAP, and suggested that a combination of biomarkers (i.e., fatty acids and stable isotopes) can 
provide additional information about food webs with more taxonomic resolution than a single 
type of biomarker alone. In line with previous fatty acid studies, we found that most species were
rich in polyunsaturated fatty acids, suggesting they could be a resource of essential fatty acids for
higher consumers. Fatty acid profiles of macroalgae grouped strongly by phylogenetic divisions, 
while stable isotopes were able to differentiate some closely-related species based on 
physiological differences. While the phylogenetic differentiation driven by fatty acids had a 
stronger influence on distinguishing Antarctic macroalgae, the added dimension of stable 
isotopes can likely make the combination of the two approaches particularly powerful in the 
application of food web studies.  
In Chapter III I used long-term data from the Oregon Marine Reserves to identify the 
distribution of target macroalgal species inside and outside of the marine reserves and determine 
135
if and how they have changed through time. Additionally, I provided methodological guidance 
on conducting more robust surveys in the future to capture changes in the community. I found 
that all targeted kelp species were observed across the marine reserves, but their presence varied 
among sites and years. I also found that monitoring data suggested declines in macroalgal cover 
over time, but this was likely attributable to methodological shifts, not true biological changes. 
These methodological shifts were primarily a result of the difficulties associated with adapting 
SCUBA surveys to open ocean conditions in Oregon, thereby limiting our ability to survey prime
subtidal kelp habitat. To address these methodological difficulties I suggested several changes to 
the monitoring program’s SCUBA surveys including: using permanent transects to detect 
changes in brown algal densities; constraining surveys to shallower (<10 m) depths to target the 
photic zone where macroalgae are found; survey during the peak production season (July-
August); and adding UAV surveys at select sites to improve tracking of macroalgal focal species.
Chapter IV moved from the whole community, to the role of a single important generalist
predator, the sunflower sea star Pycnopodia helianthoides. Our goal was to summarise the 
feeding preferences of Pycnopodia through a meta-analysis of existing literature on the wild diet 
of Pycnopodia and a cafeteria-style experiment to determine the feeding preferences of 
Pycnopodia on sea urchins and a suite of common benthic rocky habitat prey species. We 
identified only six published datasets in the literature that quantitatively described the wild diet 
of Pycnopodia. A total of 114 prey taxa were reported across all studies, with bivalves and 
urchins tending to be the primary observed prey items. This agreed with our cafeteria 
experiments that found Pycnopodia tended to prefer green and purple urchins, and mussels, 
although the quantity of each prey type consumed was highly variable. We also found that 
Pycnopodia tended to consume smaller purple and red urchins than the mean size of those 
136
species that were offered. These results suggested that Pycnopodia is a general predator but does 
exhibit a preference for urchins, supporting the idea that Pycnopodia are important in 
maintaining kelp forest health and resilience.
Finally, in Chapter V I explored the non-consumptive effects of Pycnopodia on their 
prey, the purple sea urchin Strongylocentrotus purpuratus, who is a primary kelp consumer 
across much of the West Coast of North America. Recent work has begun to quantify the 
predation rates effects (i.e., direct consumptive effects) of Pycnopodia on sea urchins that may 
lead to density-mediated indirect effects on kelp. However, the importance of non-consumptive 
effects on urchin behavior and the possible trait-mediated indirect effects of Pycnopodia on kelp 
have not been well understood. This left a critical gap in our knowledge about how these 
predators may be controlling grazer populations and, indirectly, primary production by 
macroalgae in nearshore habitats. This chapter tested the non-consumptive behavioral effects of 
Pycnopodia on S. purpuratus including grazing rates, feeding behavior, and movement of 
starved and fed urchins, the latter simulating urchin metabolic conditions within urchin barrens. I
found that the presence of a waterborne Pycnopodia cue reduced the grazing rate of fed urchins 
by 50% over short (~24 h) time scales. In contrast, starved urchins consumed kelp and did not 
exhibit an escape response in the presence of a Pycnopodia cue. This chapter highlights a trait-
mediated indirect interaction between Pycnopodia, S. purpuratus and kelp, and how the urchin 
response to a predator cue may differ based on urchin metabolic conditions. Moreover, this study
builds our understanding of the importance of the decline of Pycnopodia from sea star wasting 
disease and the positive role of Pycnopodia on kelp forest health.
Macroalgal forests are critical to the health of our coastal oceans, providing countless 
ecosystem services, both objectively quantifiable—as I have demonstrated in this dissertation—
137
and abstract. I am drawn to these forests primarily by their abstract qualities, their beauty, their 
cultural significance, and the indescribable feeling I experience when I am diving beneath their 
canopy. My greatest hope is that the objectively quantifiable qualities of macroalgal forests that I
have presented in this dissertation will help us preserve, support, and rejuvenate them now and 
for years to come, so that future generations can experience their beauty as well. 
138
REFERENCES CITED
Amsler CD, Amsler MO, McClintock JB, Baker BJ (2009) Filamentous algal endophytes in 
macrophytic Antarctic algae: Prevalence in hosts and palatability to mesoherbivores. 
Phycologia 48:324–334.
Amsler CD, Iken K, McClintock JB, Amsler MO, Peters KJ, Hubbard JM, Furrow FB, Baker BJ 
(2005) Comprehensive evaluation of the palatability and chemical defenses of subtidal 
macroalgae from the Antarctic Peninsula. Mar Ecol Prog Ser 294:141–159.
Amsler CD, Mcclintock JB, Baker BJ (2012) Palatability of living and dead detached Antarctic 
macroalgae to consumers. Antarct Sci 24:589–590.
Amsler MO, Amsler CD, Salm JL von, Aumack CF, McClintock JB, Young RM, Baker BJ 
(2013) Tolerance and sequestration of macroalgal chemical defenses by an Antarctic 
amphipod: a ‘cheater’ among mutualists. Mar Ecol Prog Ser 490:79–90.
Arts MT, Ackman RG, Holub BJ (2001) “Essential fatty acids” in aquatic ecosystems: A crucial 
link between diet and human health and evolution. Can J Fish Aquat Sci 58:122–137.
Ashton GV, Freestone AL, Duffy JE, Torchin ME, Sewall BJ, Tracy B, Albano M, Altieri AH, 
Altvater L, Bastida-Zavala R, Bortolus A, Brante A, Bravo V, Brown N, Buschmann AH,
Buskey E, Barrera RC, Cheng B, Collin R, Coutinho R, Gracia LD, Dias GM, DiBacco 
C, Flores AAV, Haddad MA, Hoffman Z, Erquiaga BI, Janiak D, Campeán AJ, Keith I, 
Leclerc J-C, Lecompte-Pérez OP, Longo GO, Matthews-Cascon H, McKenzie CH, 
Miller J, Munizaga M, Naval-Xavier LPD, Navarrete SA, Otálora C, Palomino-Alvarez 
LA, Palomo MG, Patrick C, Pegau C, Pereda SV, Rocha RM, Rumbold C, Sánchez C, 
Sanjuan-Muñoz A, Schlöder C, Schwindt E, Seemann J, Shanks A, Simoes N, Skinner L,
Suárez-Mozo NY, Thiel M, Valdivia N, Velez-Zuazo X, Vieira EA, Vildoso B, 
Wehrtmann IS, Whalen M, Wilbur L, Ruiz GM (2022) Predator control of marine 
communities increases with temperature across 115 degrees of latitude. Science 
376:1215–1219.
Assis J, Berecibar E, Claro B, Alberto F, Reed D, Raimondi P, Serrão EA (2017) Major shifts at 
the range edge of marine forests: The combined effects of climate changes and limited 
dispersal. Sci Rep 7:44348.
Aumack CF, Lowe AT, Amsler CD, Amsler MO, McClintock JB, Baker BJ (2017) Gut content, 
fatty acid, and stable isotope analyses reveal dietary sources of macroalgal-associated 
amphipods along the western Antarctic Peninsula. Polar Biol 40:1371–1384.
139
Barbeau MA, Scheibling RE (1994) Temperature effects on predation of juvenile sea scallops 
[Placopecten magellanicus (Gmelin)] by sea stars (Asterias vulgaris Verrill) and crabs 
(Cancer irroratus Say). J Exp Mar Biol Ecol 182:27–47.
Barbosa M, Fernandes F, Pereira DM, Azevedo IC, Sousa-Pinto I, Andrade PB, Valentão P 
(2020) Fatty acid patterns of the kelps Saccharina latissima, Saccorhiza polyschides and 
Laminaria ochroleuca: Influence of changing environmental conditions. Arab J Chem 
13:45–58.
Barkina MY, Pomazenkova LA, Chopenko NS, Velansky PV, Kostetsky EYa, Sanina NM 
(2019) Effect of warm acclimation rate on fatty acid composition and phase transitions of
Saccharina japonica (J.E. Areschoug) glycolipids. Vestn Tomsk Gos Univ Biol 48:135–
157.
Barnes DKA (2017) Iceberg killing fields limit huge potential for benthic blue carbon in 
Antarctic shallows. Glob Change Biol 23:2649–2659.
Barrett NS, Buxton CD, Edgar GJ (2009) Changes in invertebrate and macroalgal populations in 
Tasmanian marine reserves in the decade following protection. J Exp Mar Biol Ecol 
370:104–119.
Bates D, Mächler M, Bolker B, Walker S (2015) Fitting linear mixed-effects models using lme4. 
J Stat Softw 67:1–48.
Beardall J, Roberts S (1999) Inorganic carbon acquisition by two Antarctic macroalgae, 
Porphyra endiviifolium (Rhodophyta: Bangiales) and Palmaria decipiens (Rhodophyta: 
Palmariales). Polar Biol 21:310–315.
Beas-Luna R, Micheli F, Woodson CB, Carr M, Malone D, Torre J, Boch C, Caselle JE, 
Edwards M, Freiwald J, Hamilton SL, Hernandez A, Konar B, Kroeker KJ, Lorda J, 
Montaño‐Moctezuma G, Torres‐Moye G (2020) Geographic variation in responses of 
kelp forest communities of the California Current to recent climatic changes. Glob 
Change Biol 26:6457–6473.
Becker OS code by RA, Minka ARWR version by RBE by TP, Deckmyn A (2022) Maps: Draw 
geographical maps.
Becker S, Graeve M, Bischof K (2010) Photosynthesis and lipid composition of the Antarctic 
endemic rhodophyte Palmaria decipiens: effects of changing light and temperature 
levels. Polar Biol 33:945–955.
Beisner B, Haydon D, Cuddington K (2003) Alternative stable states in ecology. Front Ecol 
Environ 1:376–382.
140
Belghit I, Rasinger JD, Heesch S, Biancarosa I, Liland N, Torstensen B, Waagbø R, Lock E-J, 
Bruckner CG (2017) In-depth metabolic profiling of marine macroalgae confirms strong 
biochemical differences between brown, red and green algae. Algal Res 26:240–249.
Belleza D, Kawabata Y, Toda T, Nishihara G (2021) Effects of dead conspecifics, hunger states, 
and seasons on the foraging behavior of the purple urchin Heliocidaris crassispina. Mar 
Ecol Prog Ser 664:133–148.
Berneira L, da Silva C, Poletti T, Ritter M, dos Santos M, Colepicolo P, de Pereira CMP (2020) 
Evaluation of the volatile composition and fatty acid profile of seven Antarctic 
macroalgae. J Appl Phycol 32:3319–3329.
Bernstein BB, Williams BE, Mann KH (1981) The role of behavioral responses to predators in 
modifying urchins’ (Strongylocentrotus droebachiensis) destructive grazing and seasonal 
foraging patterns. Mar Biol 63:39–49.
Bodkin JL (2015) Chapter 3 - Historic and contemporary status of sea otters in the North Pacific.
In: Sea Otter Conservation. Larson SE, Bodkin JL, VanBlaricom GR (eds) Academic 
Press, Boston, p 43–61
Boecklen WJ, Yarnes CT, Cook BA, James AC (2011) On the use of stable isotopes in trophic 
ecology. Annu Rev Ecol Evol Syst 42:411–440.
Bograd SJ, Mendelssohn R, Schwing FB, Miller AJ (2005) Spatial heterogeneity of sea surface 
temperature trends in the Gulf of Alaska. Atmosphere-Ocean 43:241–247.
Bolker B, Holyoak M, Křivan V, Rowe L, Schmitz O (2003) Connecting theoretical and 
empirical studies of trait-mediated interactions. Ecology 84:1101–1114.
Bonaviri C, Graham M, Gianguzza P, Shears NT (2017) Warmer temperatures reduce the 
influence of an important keystone predator. J Anim Ecol 86:490–500.
Bourdeau PE, Bach MT, Peacor SD (2016) Predator presence dramatically reduces copepod 
abundance through condition-mediated non-consumptive effects. Freshw Biol 61:1020–
1031.
Bracken MES, Nielsen KJ (2004) Diversity of intertidal macroalgae increases with nitrogen 
loading by invertebrates. Ecology 85:2828–2836.
Brett MT, Eisenlord ME, Galloway AWE (2016) Using multiple tracers and directly accounting 
for trophic modification improves dietary mixing-model performance. Ecosphere 
7:e01440.
141
Brewer R, Konar B (2005) Chemosensory responses and foraging behavior of the seastar 
Pycnopodia helianthoides. Mar Biol 147:789–795.
Bromaghin JF (2017) QFASAR: Quantitative fatty acid signature analysis with R. Methods Ecol 
Evol 8:1158–1162.
Brown D, Christian W, Hanson RM (2021) Tracker: Video analysis and modeling tool.
Brown JS, Laundré JW, Gurung M (1999) The ecology of fear: Optimal foraging, game theory, 
and trophic interactions. J Mammal 80:385–399.
Brown MB, Edwards MS, Kim KY (2014) Effects of climate change on the physiology of giant 
kelp, Macrocystis pyrifera, and grazing by purple urchin, Strongylocentrotus purpuratus. 
ALGAE 29:203–215.
Budge SM, Iverson SJ, Koopman HN (2006) Studying trophic ecology in marine ecosystems 
using fatty acids: A primer on analysis and interpretation. Mar Mammal Sci 22:759–801.
Burt JM, Tinker MT, Okamoto DK, Demes KW, Holmes K, Salomon AK (2018) Sudden 
collapse of a mesopredator reveals its complementary role in mediating rocky reef regime
shifts. Proc R Soc B Biol Sci 285:20180553.
Byrne ME, Vaudo JJ, Harvey GCMcN, Johnston MW, Wetherbee BM, Shivji M (2019) 
Behavioral response of a mobile marine predator to environmental variables differs 
across ecoregions. Ecography 42:1569–1578.
Byrnes J, Reed D (2018) KEEN protocol manual v0.9.5.
Byrnes J, Stachowicz JJ, Hultgren KM, Hughes AR, Olyarnik SV, Thornber CS (2006) Predator 
diversity strengthens trophic cascades in kelp forests by modifying herbivore behaviour. 
Ecol Lett 9:61–71.
Byrnes JE, Reed DC, Cardinale BJ, Cavanaugh KC, Holbrook SJ, Scmidtt RJ (2011) Climate-
driven increases in storm frequency simplify kelp forest food webs. Glob Change Biol 
17:2513–2524.
Calenge C (2006) The package adehabitat for the R software: Tool for the analysis of space and 
habitat use by animals. Ecol Model 197:1035.
Campbell AC, Coppard S, D’Abreo C, Tudor-Thomas R (2001) Escape and aggregation 
responses of three echinoderms to conspecific stimuli. Biol Bull 201:175–185.
142
Cardona L, Lloret-Lloret E, Moles J, Avila C (2021) Latitudinal changes in the trophic structure 
of benthic coastal food webs along the Antarctic Peninsula. Mar Environ Res 
167:105290.
Carr MH, Caselle JE, Koehn KD, Malone DP (2020) PISCO kelp forest community surveys.
Casaux R, Barrera-Oro E (2013) Dietary overlap in inshore notothenioid fish from the Danco 
Coast, western Antarctic Peninsula. Polar Res 32:21319.
Caut S, Angulo E, Courchamp F (2009) Variation in discrimination factors (Δ 15 N and Δ 13 C): 
The effect of diet isotopic values and applications for diet reconstruction. J Appl Ecol 
46:443–453.
Chapman RL, Bailey JC, Waters DA (2012) Macroalgal phylogeny, Chapter 20. In: Molecular 
Approaches to the Study of the Ocean. Cooksey K (ed) Springer Science & Business 
Media, p 389
Chivers DP, Dixson DL, White JR, McCormick MI, Ferrari MCO (2013) Degradation of 
chemical alarm cues and assessment of risk throughout the day. Ecol Evol 3:3925–3934.
Cordone G, Marina TI, Salinas V, Doyle SR, Saravia LA, Momo FR (2018) Effects of 
macroalgae loss in an Antarctic marine food web: applying extinction thresholds to food 
web studies. PeerJ 6:e5531.
Cramer M, Blanc AL, Flagel D, Driscoll S (2020) Differentiating diverse diets: A novel 
approach to analyzing cafeteria feeding trials. Preprints.
Cruz-Rivera E, Hay ME (2000) Can quantity replace quality? Food choice, compensatory 
feeding, and fitness of marine mesograzers. Ecology 81:201–219.
Dalsgaard J, St. John M, Kattner G, Müller-Navarra D, Hagen W (2003) Fatty acid trophic 
markers in the pelagic marine environment. In: Advances in Marine Biology. Elsevier, p 
225–340
Dare PJ, Edwards DB (1975) Changes in flesh weight and biochemical composition of mussels 
(Mytilus edulis L.) in the Conwy Estuary, North Wales. J Exp Mar Biol Ecol 18:89–97.
Dauby P, Nyssen F, Broyer CD (2003) Amphipods as food sources for higher trophic levels in 
the Southern Ocean: A synthesis. In: Antarctic Biology in a Global Context. Huiskes A, 
Gieskes W, Rozema J, Schorno R, van der Vies S, Antz W (eds) Backhuys Publishers, 
Leiden, The Netherlands, p 29–134
Dayton PK (1985) Ecology of kelp communities. Annu Rev Ecol Syst 16:215–245.
143
Dehaudt B, Nguyen M, Vadlamudi A, Blumstein DT (2019) Giant clams discriminate threats 
along a risk gradient and display varying habituation rates to different stimuli. Ethology 
125:392–398.
DeNiro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in animals. 
Geochim Cosmochim Acta 42:495–506.
Deregibus D, Quartino ML, Campana GL, Momo FR, Wiencke C, Zacher K (2016) 
Photosynthetic light requirements and vertical distribution of macroalgae in newly ice-
free areas in Potter Cove, South Shetland Islands, Antarctica. Polar Biol 39:153–166.
Dethier MN, Sosik E, Galloway AWE, Duggins DO, Simenstad CA (2013) Addressing 
assumptions: Variation in stable isotopes and fatty acids of marine macrophytes can 
confound conclusions of food web studies. Mar Ecol Prog Ser 478:1–14.
Dolinar D, Edwards M (2021) The metabolic depression and revival of purple urchins 
(Strongylocentrotus purpuratus) in response to macroalgal availability. J Exp Mar Biol 
Ecol 545:151646.
Douglass JG, Duffy JE, Bruno JF (2008) Herbivore and predator diversity interactively affect 
ecosystem properties in an experimental marine community: Interacting changes in 
predator and prey diversity. Ecol Lett 11:598–608.
Draper AM, Weissburg MJ (2019) Impacts of global warming and elevated CO2 on sensory 
behavior in predator-prey interactions: A review and synthesis. Front Ecol Evol 7:72.
Duarte CM (2017) Reviews and syntheses: Hidden forests, the role of vegetated coastal habitats 
in the ocean carbon budget. Biogeosciences 14:301–310.
Duggins D, Eckman J, Siddon C, Klinger T (2001) Interactive roles of mesograzers and current 
flow in survival of kelps. Mar Ecol Prog Ser 223:143–155.
Duggins DO (1983) Starfish predation and the creation of mosaic patterns in a kelp-dominated 
community. Ecology 64:1610–1619.
Dunstan G, Brown M, Volkman J (2005) Cryptophyceae and Rhodophyceae; chemotaxonomy, 
phylogeny, and application. Phytochemistry 66:2557–2570.
Dunton KH (2001) Δ15N and δ13C measurements of Antarctic Peninsula fauna: Trophic 
relationships and assimilation of benthic seaweeds. Am Nat 41:99–112.
144
Eddy TD, Bernhardt JR, Blanchard JL, Cheung WWL, Colléter M, du Pontavice H, Fulton EA, 
Gascuel D, Kearney KA, Petrik CM, Roy T, Rykaczewski RR, Selden R, Stock CA, 
Wabnitz CCC, Watson RA (2021) Energy flow through marine ecosystems: Confronting 
transfer efficiency. Trends Ecol Evol 36:76–86.
Eger AM, Marzinelli EM, Beas-Luna R, Blain CO, Blamey LK, Byrnes JEK, Carnell PE, Choi 
CG, Hessing-Lewis M, Kim KY, Kumagai NH, Lorda J, Moore P, Nakamura Y, Pérez-
Matus A, Pontier O, Smale D, Steinberg PD, Vergés A (2023) The value of ecosystem 
services in global marine kelp forests. Nat Commun 14:1894.
Eisaguirre JH, Eisaguirre JM, Davis K, Carlson PM, Gaines SD, Caselle JE (2020) Trophic 
redundancy and predator size class structure drive differences in kelp forest ecosystem 
dynamics. Ecology 101:e02993.
Elahi R, Sebens KP (2013) Experimental removal and recovery of subtidal grazers highlights the
importance of functional redundancy and temporal context. PLoS ONE 8:e78969.
Ellingsen KE, Anderson MJ, Shackell NL, Tveraa T, Yoccoz NG, Frank KT (2015) The role of a
dominant predator in shaping biodiversity over space and time in a marine ecosystem. J 
Anim Ecol 84:1242–1252.
Elliott JM (2006) Prey switching in Rhyacophila dorsalis (Trichoptera) alters with larval instar. 
Freshw Biol 51:913–924.
Estes JA, Duggins DO (1995) Sea otters and kelp forests in Alaska: Generality and variation in a 
community ecological paradigm. Ecol Monogr 65:75–100.
Filbee-Dexter K, Scheibling RE (2014) Sea urchin barrens as alternative stable states of 
collapsed kelp ecosystems. Mar Ecol Prog Ser 495:1–25.
Filbee-Dexter K, Scheibling RE (2016) Spatial patterns and predictors of drift algal subsidy in 
deep subtidal environments. Estuaries Coasts 39:1724–1734.
Fischer G, Wiencke C (1992) Stable carbon isotope composition, depth distribution and fate of 
macroalgae from the Antarctic Peninsula region. Polar Biol 12:341–348.
France R (1995) Carbon-13 enrichment in benthic compared to planktonic algae: Foodweb 
implications. Mar Ecol Prog Ser 124:307–312.
Freas CA, Cheng K (2022) Neuroecology beyond the brain: learning in Echinodermata. Learn 
Behav 50:20–36.
Freeman A (2006) Size-dependent trait-mediated indirect interactions among sea urchin 
herbivores. Behav Ecol 17:182–187.
145
Frölicher TL, Fischer EM, Gruber N (2018) Marine heatwaves under global warming. Nature 
560:360–364.
Gabara SS, Konar BH, Edwards MS (2021) Biodiversity loss leads to reductions in community‐
wide trophic complexity. Ecosphere 12.
Galloway AWE, Britton-Simmons KH, Duggins DO, Gabrielson PW, Brett MT (2012) Fatty 
acid signatures differentiate marine macrophytes at ordinal and family ranks. J Phycol 
48:956–965.
Galloway AWE, Budge SM (2020) The critical importance of experimentation in biomarker-
based trophic ecology. Philos Trans R Soc B Biol Sci 375:20190638.
Galloway AWE, Gravem SA, Kobelt JN, Heady WN, Okamoto DK, Sivitilli DM, Saccomanno 
VR, Hodin J, Whippo R (2023) Sunflower sea star predation on urchins can facilitate 
kelp forest recovery. Proc R Soc B Biol Sci 290:20221897.
Galloway AWE, Winder M (2015) Partitioning the relative importance of phylogeny and 
environmental conditions on phytoplankton fatty acids. PLOS ONE 10:e0130053.
Garnier S, Ross N, Rudis B, Filipovic-Pierucci A, Galili T, Greenwell B, Sievert C, Harris DJ, 
Sciaini M, Chen JJ (2021) Viridis - Colorblind-Friendly Color Maps for R: CRAN 
release v0.6.2.
Gasperini S, Bonacchi A, Bartolommei P, Manzo E, Cozzolino R (2018) Seasonal cravings: 
plant food preferences of syntopic small mammals. Ethol Ecol Evol 30:12–25.
Gaymer C, Himmelman J, Johnson L (2002) Effect of intra- and interspecific interactions on the 
feeding behavior of two subtidal sea stars. Mar Ecol Prog Ser 232:149–162.
Graeve M, Kattner G, Wiencke C, Karsten U (2002) Fatty acid composition of Arctic and 
Antarctic macroalgae: Indicator of phylogenetic and trophic relationships. Mar Ecol Prog 
Ser 231:67–74.
Gravem S, Heady W, Saccomanno VR, Alvstad KF, Gehman AL, Frierson TN, Hamilton SL 
(2021) Pycnopodia helianthoides. IUCN redlist of threatened species 2021. IUCN.
Gravem SA, Morgan SG (2019) Trait-mediated indirect effects in a natural tidepool system. Mar 
Biol 166:23.
Green L (2012) Refuge availability increases kelp consumption by purple sea urchins exposed to 
predation risk cue. Aquat Biol 17:141–144.
146
Grotti M, Soggia F, Abelmoschi ML, Rivaro P, Magi E, Frache R (2001) Temporal distribution 
of trace metals in Antarctic coastal waters. Mar Chem 76:189–209.
Guerrero AI, Rogers TL (2020) Evaluating the performance of the Bayesian mixing tool 
MixSIAR with fatty acid data for quantitative estimation of diet. Sci Rep 10:20780.
Guest MA, Hirst AJ, Nichols PD, Frusher SD (2010) Multi-scale spatial variation in stable 
isotope and fatty acid profiles amongst temperate reef species: Implications for design 
and interpretation of trophic studies. Mar Ecol Prog Ser 410:25–41.
Hadfield MG, Paul VJ (2001) Natural chemical cues for settlement and metamorphosis of 
marine-invertebrate larvae.
Haggerty MB, Anderson TW, Long JD (2018) Fish predators reduce kelp frond loss via a trait-
mediated trophic cascade. Ecology 99:1574–1583.
Halpern BS (2003) The impact of marine reserves: Do reserves work and does reserve size 
matter? Ecol Appl 13:117–137.
Hamel J-F, Jobson S, Caulier G, Mercier A (2021) Evidence of anticipatory immune and 
hormonal responses to predation risk in an echinoderm. Sci Rep 11:10691.
Hamilton SL, Bell TW, Watson JR, Grorud-Colvert KA, Menge BA (2020) Remote sensing: 
Generation of long-term kelp bed data sets for evaluation of impacts of climatic variation.
Ecology 101.
Hamilton SL, Caselle JE (2015) Exploitation and recovery of a sea urchin predator has 
implications for the resilience of southern California kelp forests. Proc R Soc B Biol Sci 
282:20141817.
Hamilton SL, Saccomanno VR, Heady WN, Gehman AL, Lonhart SI, Beas-Luna R, Francis FT, 
Lee L, Rogers-Bennett L, Salomon AK, Gravem SA (2021) Disease-driven mass 
mortality event leads to widespread extirpation and variable recovery potential of a 
marine predator across the eastern Pacific. Proc R Soc B Biol Sci 288:20211195.
Hanson CE, Hyndes GA, Wang SF (2010) Differentiation of benthic marine primary producers 
using stable isotopes and fatty acids: Implications to food web studies. Aquat Bot 
93:114–122.
Harding APC, Scheibling RE (2015) Feed or flee: Effect of a predation-risk cue on sea urchin 
foraging activity. J Exp Mar Biol Ecol 466:59–69.
Harley CDG, Anderson KM, Demes KW, Jorve JP, Kordas RL, Coyle TA, Graham MH (2012) 
Effects of climate change on global seaweed communities. J Phycol 48:1064–1078.
147
Harvell CD, Montecino-Latorre D, Caldwell JM, Burt JM, Bosley K, Keller A, Heron SF, 
Salomon AK, Lee L, Pontier O, Pattengill-Semmens C, Gaydos JK (2019) Disease 
epidemic and a marine heat wave are associated with the continental-scale collapse of a 
pivotal predator Pycnopodia helianthoides. Sci Adv:9.
Hawlena D, Schmitz OJ (2010) Herbivore physiological response to predation risk and 
implications for ecosystem nutrient dynamics. Proc Natl Acad Sci 107:15503–15507.
Heiser S, Amsler CD, Brothers CJ, Amsler MO, Shilling AJ, Bozarth L, Davis CB, McClintock 
JB, Baker BJ (2022) Who cares more about chemical defenses — the macroalgal 
producer or its main grazer? J Chem Ecol 48:416–430.
Heithaus MR, Wirsing AJ, Burkholder D, Thomson J, Dill LM (2009) Towards a predictive 
framework for predator risk effects: The interaction of landscape features and prey escape
tactics. J Anim Ecol 78:556–562.
Hernández-Carmona G, Carrillo-Domínguez S, Arvizu-Higuera DL, Rodríguez-Montesinos YE, 
Murillo-Álvarez JI, Muñoz-Ochoa M, Castillo-Domínguez RM (2009) Monthly variation
in the chemical composition of Eisenia arborea J.E. Areschoug. J Appl Phycol 21:607–
616.
Herrlinger TJ (1983) A thesis presented to the faculty of Moss Landing Marine Laboratories San 
Jose State University. San Jose State University, Moss Landing Marine Laboratories
Hessing-Lewis ML, Hacker SD (2013) Upwelling-influence, macroalgal blooms, and seagrass 
production; temporal trends from latitudinal and local scales in northeast Pacific 
estuaries. Limnol Oceanogr 58:1103–1112.
Hickey BM, Royer TC (2010) California and Alaska currents. In: Ocean Currents. Steele JH, 
Thorpe SA, Turekian KK (eds) Academic Press, p 73–83
Hodin J, Pearson-Lund A, Anteau FP, Kitaeff P, Cefalu S (2021) Progress toward complete life-
cycle Culturing of the endangered sunflower star, Pycnopodia helianthoides. Biol Bull 
241:243–258.
Huang YM, McClintock JB, Amsler CD, Peters KJ, Baker BJ (2006) Feeding rates of common 
Antarctic gammarid amphipods on ecologically important sympatric macroalgae. J Exp 
Mar Biol Ecol 329:55–65.
Hughes BB (2010) Variable effects of a kelp foundation species on rocky intertidal diversity and 
species interactions in central California. J Exp Mar Biol Ecol 393:90–99.
Hughes RN, Croy MI (1993) An experimental analysis of frequency-dependent predation 
(switching) in the 15-spined stickleback, Spinachia spinachia. J Anim Ecol 62:341.
148
Iannone R, Cheng J, Schloerke B (2022) Gt: Easily create presentation-ready display tables.
Iken K (1999) Feeding ecology of the Antarctic herbivorous gastropod Laevilacunaria 
antarctica Martens. J Exp Mar Biol Ecol 236:133–148.
Iken K, Amsler CD, Hubbard JM, McClintock JB, Baker BJ (2007) Allocation patterns of 
phlorotannins in Antarctic brown algae. Phycologia 46:386–395.
Iken K, Barrera-Oro ER, Quartino ML, Casaux RJ, Brey T (1997) Grazing by the Antarctic fish 
Notothenia coriiceps: evidence for selective feeding on macroalgae. Antarct Sci 9:386–
391.
Iverson SJ, Field C, Bowen WD, Blanchard W (2004) Quantitative fatty acid signature analysis: 
A new method of estimating predator diets. Ecol Monogr 74:211–235.
Ives AR, Cardinale BJ, Snyder WE (2005) A synthesis of subdisciplines: Predator–prey 
interactions, and biodiversity and ecosystem functioning. Ecol Lett 8:102–116.
Kain JM (1987) Patterns of relative growth in Nereocystis luetkeana (Phaeophyta). J Phycol 
23:181–187.
Kalinkat G, Rall BC, Vucic-Pestic O, Brose U (2011) The allometry of prey preferences. PLOS 
ONE 6:e25937.
Kassambara A, Mundt F (2020) Factoextra: Extract and visualize the results of multivariate data 
analyses.
Kats LB, Dill LM (1998) The scent of death: Chemosensory assessment of predation risk by prey
animals. Écoscience 5:361–394.
Kelly J, Scheibling R (2012) Fatty acids as dietary tracers in benthic food webs. Mar Ecol Prog 
Ser 446:1–22.
Kempf A, Floeter J, Temming A (2008) Predator–prey overlap induced Holling type III 
functional response in the North Sea fish assemblage. Mar Ecol Prog Ser 367:295–308.
Khotimchenko SV (1998) Fatty acids of brown algae from the Russian far east. Phytochemistry 
49:2363–2369.
Khotimchenko SV, Vaskovsky VE, Titlyanova TV (2002) Fatty acids of marine algae from the 
Pacific Coast of North California. Bot Mar 45:17–22.
Kim JK, Kraemer GP, Yarish C (2013) Emersion induces nitrogen release and alteration of 
nitrogen metabolism in the intertidal genus Porphyra. PLOS ONE 8:e69961.
149
Kissling WD, Schleuning M (2015) Multispecies interactions across trophic levels at 
macroscales: Retrospective and future directions. Ecography 38:346–357.
Knight CJ, Dunn RP, Long JD (2022) Conspecific cues, not starvation, mediate barren urchin 
response to predation risk. Oecologia.
Kohl KD, Coogan SCP, Raubenheimer D (2015) Do wild carnivores forage for prey or for 
nutrients?: Evidence for nutrient-specific foraging in vertebrate predators. BioEssays 
37:701–709.
Krause-Jensen D, Duarte CM (2016) Substantial role of macroalgae in marine carbon 
sequestration. Nat Geosci 9:737–742.
Krieg K, Menge B, Lubchenco J (2019) Field guide to Oregon’s rocky intertidal.
Kriegisch N, Reeves SE, Flukes EB, Johnson CR, Ling SD (2019) Drift-kelp suppresses foraging
movement of overgrazing sea urchins. Oecologia 190:665–677.
Krumhansl KA, Lauzon-Guay J-S, Scheibling RE (2014) Modeling effects of climate change and
phase shifts on detrital production of a kelp bed. Ecology 95:763–774.
Krumhansl KA, Okamoto DK, Rassweiler A, Novak M, Bolton JJ, Cavanaugh KC, Connell SD, 
Johnson CR, Konar B, Ling SD, Micheli F, Norderhaug KM, Pérez-Matus A, Sousa-
Pinto I, Reed DC, Salomon AK, Shears NT, Wernberg T, Anderson RJ, Barrett NS, 
Buschmann AH, Carr MH, Caselle JE, Derrien-Courtel S, Edgar GJ, Edwards M, Estes 
JA, Goodwin C, Kenner MC, Kushner DJ, Moy FE, Nunn J, Steneck RS, Vásquez J, 
Watson J, Witman JD, Byrnes JEK (2016) Global patterns of kelp forest change over the 
past half-century. Proc Natl Acad Sci 113:13785–13790.
Krumhansl KA, Scheibling RE (2012) Production and fate of kelp detritus. Mar Ecol Prog Ser 
467:281–302.
Kumari P, Bijo AJ, Mantri VA, Reddy CRK, Jha B (2013) Fatty acid profiling of tropical marine
macroalgae: An analysis from chemotaxonomic and nutritional perspectives. 
Phytochemistry 86:44–56.
Lamy T, Reed DC, Rassweiler A, Siegel DA, Kui L, Bell TW, Simons RD, Miller RJ (2018) 
Scale-specific drivers of kelp forest communities. Oecologia 186:217–233.
Layman CA, Arrington DA, Montaña CG, Post DM (2007) Can stable isotope ratios provide for 
community-wide measures of trophic structure? Ecology 88:42–48.
Lee LC, Watson JC, Trebilco R, Salomon AK (2016) Indirect effects and prey behavior mediate 
interactions between an endangered prey and recovering predator. Ecosphere 7:e01604.
150
Lefcheck JS, Hughes BB, Johnson AJ, Pfirrmann BW, Rasher DB, Smyth AR, Williams BL, 
Beck MW, Orth RJ (2019) Are coastal habitats important nurseries? A meta-analysis. 
Conserv Lett 0:e12645.
Lemesle S, Erraud A, Mussio I, Rusig A-M, Claquin P (2016) Dynamics of δ15N isotopic 
signatures of different intertidal macroalgal species: Assessment of bioindicators of N 
sources in coastal areas. Mar Pollut Bull 110:470–483.
Liyana-Pathirana C, Shahidi F, Whittick A (2002) The effect of an artificial diet on the 
biochemical composition of the gonads of the sea urchin (Strongylocentrotus 
droebachiensis). Food Chem 79:461–472.
Lowe AT, Whippo R, Galloway AWE, Britton-Simmons KH, Dethier MN (2015) Sedentary 
urchins influence benthic community composition below the macroalgal zone. Mar Ecol 
36:129–140.
Mackey AP, Hyndes GA, Carvalho MC, Eyre BD (2015) Physical and biogeochemical correlates
of spatio-temporal variation in the δ13C of marine macroalgae. Estuar Coast Shelf Sci 
157:7–18.
Mann KH (1973) Seaweeds: Their productivity and strategy for growth. Science 182:975–981.
Manzur T, Barahona M, Navarrete SA (2010) Ontogenetic changes in habitat use and diet of the 
sea-star Heliaster helianthus on the coast of central Chile. J Mar Biol Assoc U K 90:537–
546.
Marconi M, Giordano M, Raven JA (2011) Impact of taxonomy, geography, and depth on δ13C 
and δ15N variation in a large collection of macroalgae. J Phycol 47:1023–1035.
Marliave JB (1987) The life history and captive reproduction of the wolf-eel Anarrhichthys 
ocellatus at the Vancouver Public Aquarium. Int Zoo Yearb 26:70–81.
Mauzey KP, Birkeland C, Dayton PK (1968) Feeding behavior of asteroids and escape responses
of their prey in the Puget Sound region. Ecology 49:603–619.
Mauzole YL, Torres HS, Fu L ‐L. (2020) Patterns and dynamics of SST fronts in the California 
Current system. J Geophys Res Oceans 125.
McClintock JB, Lawrence JM (1985) Characteristics of foraging in the soft-bottom benthic 
starfish Luidia clathrata (echinodermata: Asteroidea): prey selectivity, switching 
behavior, functional responses and movement patterns. Oecologia 66:291–298.
McGaw IJ, Twitchit TA (2012) Specific dynamic action in the sunflower star, Pycnopodia 
helianthoides. Comp Biochem Physiol A Mol Integr Physiol 161:287–295.
151
McPherson ML, Finger DJI, Houskeeper HF, Bell TW, Carr MH, Rogers-Bennett L, Kudela RM
(2021) Large-scale shift in the structure of a kelp forest ecosystem co-occurs with an 
epizootic and marine heatwave. Commun Biol 4:1–9.
Medrano A, Linares C, Aspillaga E, Capdevila P, Montero-Serra I, Pagès-Escolà M, Zabala M, 
Hereu B (2020) Long-term monitoring of temperate macroalgal assemblages inside and 
outside a no take marine reserve. Mar Environ Res 153:104826.
Menge B, Farrell TM, Olson AM, van Tamelen P, Turner T (1993) Algal recruitment and the 
maintenance of a plant mosaic in the low intertidal region on the Oregon coast. J Exp 
Mar Biol Ecol 170:91–116.
Menge BA, Allison GW, Blanchette CA, Farrell TM, Olson AM, Turner TA, van Tamelen P 
(2005) Stasis or kinesis? Hidden dynamics of a rocky intertidal macrophyte mosaic 
revealed by a spatially explicit approach. J Exp Mar Biol Ecol 314:3–39.
Moitoza DJ, Phillips DW (1979) Prey defense, predator preference, and nonrandom diet: The 
interactions between Pycnopodia helianthoides and two species of sea urchins. Mar Biol 
53:299–304.
Molloy PP, McLean IB, Côté IM (2009) Effects of marine reserve age on fish populations: A 
global meta-analysis. J Appl Ecol 46:743–751.
Montgomery EM (2014) Predicting crawling speed relative to mass in sea stars. J Exp Mar Biol 
Ecol 458:27–33.
Müller-Navarra DC, Brett MT, Liston AM, Goldman CR (2000) A highly unsaturated fatty acid 
predicts carbon transfer between primary producers and consumers. Nature 403:74–77.
Murdoch WW (1969) Switching in general predators: Experiments on predator specificity and 
stability of prey populations. Ecol Monogr 39:335–354.
Mystikou A, Peters AF, Asensi AO, Fletcher KI, Brickle P, van West P, Convey P, Küpper FC 
(2014) Seaweed biodiversity in the south-western Antarctic Peninsula: Surveying 
macroalgal community composition in the Adelaide Island/Marguerite Bay region over a 
35-year time span. Polar Biol 37:1607–1619.
Newell RIE (2004) Ecosystem influences of natural and cultivated populations of suspension-
feeding bivalve molluscs: A review. J Shellfish Res 23:51.
Nichols KD, Segui L, Hovel KA (2015) Effects of predators on sea urchin density and habitat 
use in a southern California kelp forest. Mar Biol 162:1227–1237.
152
Nishizaki MT, Ackerman JD (2007) Juvenile–adult associations in sea urchins 
(Strongylocentrotus franciscanus and S. droebachiensis): Protection from predation and 
hydrodynamics in S. franciscanus. Mar Biol 151:135–145.
Norkko A, Thrush SF, Cummings VJ, Funnell GA, Schwarz A-M, Andrew NL, Hawes I (2004) 
Ecological role of Phyllophora antarctica drift accumulations in coastal soft-sediment 
communities of McMurdo Sound, Antarctica. Polar Biol 27:482–494.
O’Connor MI, Gonzalez A, Byrnes JEK, Cardinale BJ, Duffy JE, Gamfeldt L, Griffin JN, 
Hooper D, Hungate BA, Paquette A, Thompson PL, Dee LE, Dolan KL (2017) A general
biodiversity-function relationship is mediated by trophic level. Oikos 126:18–31.
Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O’Hara 
RB, Simpson GL, Solymos P, Henry M, Stevens M, Szoecs E, Wagner H (2018) Vegan: 
Community ecology package.
Oliveira MC, Pellizzari F, Medeiros AS, Yokoya NS (2020) Diversity of Antarctic seaweeds. In: 
Antarctic Seaweeds: Diversity, Adaptation and Ecosystem Services. Gómez I, Huovinen 
P (eds) Springer International Publishing, Cham, p 23–42
Ostfeld RS (1982) Foraging strategies and prey switching in the California sea otter. Oecologia 
53:170–178.
Pagès JF, Bartumeus F, Romero J, Alcoverro T (2021) The scent of fear makes sea urchins go 
ballistic. Mov Ecol 9:50.
Paine RT (1966) Food web complexity and species diversity. Am Nat 100:65–75.
Parnell PE, Fumo JT, Lennert-Cody CE, Schroeter SC, Dayton PK (2017) Sea urchin behavior in
a Southern California kelp forest: Food, fear, behavioral niches, and scaling up individual
behavior. J Shellfish Res 36:529-.
Parrish CC (2009) Essential fatty acids in aquatic food webs. In: Lipids in Aquatic Ecosystems. 
Kainz M, Brett MT, Arts MT (eds) Springer, New York, NY, p 309–326
Paul AJ, Feder HM (1975) The food of the sea star Pycnopodia helianthoides (Brandt) in Prince 
William Sound, Alaska. Ophelia 14:15–22.
Pellizzari F, Silva MC, Silva EM, Medeiros A, Oliveira MC, Yokoya NS, Pupo D, Rosa LH, 
Colepicolo P (2017) Diversity and spatial distribution of seaweeds in the South Shetland 
Islands, Antarctica: An updated database for environmental monitoring under climate 
change scenarios. Polar Biol 40:1671–1685.
153
Peña TS, Watson JR, González-Guzmán LI, Keitt TH (2017) Step-wise drops in modularity and 
the fragmentation of exploited marine metapopulations. Landsc Ecol 32:1643–1656.
Peters KJ, Amsler CD, Amsler MO, McClintock JB, Dunbar RB, Baker BJ (2005) A 
comparative analysis of the nutritional and elemental composition of macroalgae from 
the western Antarctic Peninsula. Phycologia 44:453–463.
Peterson BJ (1999) Stable isotopes as tracers of organic matter input and transfer in benthic food 
webs: A review. Acta Oecologica 20:479–487.
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst 18:293–
320.
Pfister CA, Berry HD, Mumford T (2017) The dynamics of kelp forests in the Northeast Pacific 
Ocean and the relationship with environmental drivers. J Ecol 106:1520–1533.
Pfister CA, Berry HD, Mumford T (2018) The dynamics of kelp forests in the Northeast Pacific 
Ocean and the relationship with environmental drivers. J Ecol 106:1520–1533.
Polis GA, Anderson WB, Holt RD (1997) Toward an integration of landscape and food web 
ecology: The dynamics of spatially subsidized food webs. Annu Rev Ecol Syst 28:289–
316.
Post DM (2002) Using stable isotopes to estimate trophic position: Models, methods, and 
assumptions. Ecology 83:703–718.
Preisser EL, Bolnick DI, Benard MF (2005) Scared to death? The effects of intimidation and 
consumption in predator-prey interactions. Ecology 86:501–509.
Prokopenko CM, Avgar T, Ford A, Vander Wal E (2023) Optimal prey switching: Predator 
foraging costs provide a mechanism for functional responses in multi‐prey systems. 
Ecology 104:e3928.
Pyke GH, Pulliam HR, Charnov EL (1977) Optimal foraging: A selective review of theory and 
tests. Q Rev Biol 52:137–154.
Quartino ML, Boraso de Zaixso AL (2008) Summer macroalgal biomass in Potter Cove, South 
Shetland Islands, Antarctica: its production and flux to the ecosystem. Polar Biol 31:281–
294.
Quartino ML, Deregibus D, Campana GL, Latorre GEJ, Momo FR (2013) Evidence of 
macroalgal colonization on newly ice-free areas following glacial retreat in Potter Cove 
(South Shetland Islands), Antarctica. PLOS ONE 8:e58223.
154
R Core Team (2022) R: A language and environment for statistical computing.
Raven JA, Ball LA, Beardall J, Giordano M, Maberly SC (2005) Algae lacking carbon-
concentrating mechanisms. Can J Bot 83:879–890.
Raven JA, Giordano M, Beardall J, Maberly SC (2012) Algal evolution in relation to 
atmospheric CO2: Carboxylases, carbon-concentrating mechanisms and carbon oxidation
cycles. Philos Trans R Soc B Biol Sci 367:493–507.
Raven JA, Johnston AM, Kübler JE, Korb R, McInroy SG, Handley LL, Scrimgeour CM, 
Walker DI, Beardall J, Vanderklift M, Fredriksen S, Dunton KH (2002) Mechanistic 
interpretation of carbon isotope discrimination by marine macroalgae and seagrasses. 
Funct Plant Biol 29:355–378.
Reichardt W, Dieckmann G (1985) Kinetics and trophic role of bacterial degradation of macro-
algae in Antarctic coastal waters. In: Antarctic Nutrient Cycles and Food Webs. Siegfried
WR, Condy PR, Laws RM (eds) Springer, Berlin, Heidelberg, p 115–122
Ripple WJ, Estes JA, Schmitz OJ, Constant V, Kaylor MJ, Lenz A, Motley JL, Self KE, Taylor 
DS, Wolf C (2016) What is a trophic cascade? Trends Ecol Evol 31:842–849.
Rochette R, Arsenault DJ, Justome B, Himmelman JH (1998) Chemically-mediated predator-
recognition learning in a marine gastropod. Écoscience 5:353–360.
Rodgers AR (1990) Evaluating preference in laboratory studies of diet selection. Can J Zool 
68:188–190.
Rogers-Bennett L, Catton CA (2019) Marine heat wave and multiple stressors tip bull kelp forest
to sea urchin barrens. Sci Rep 9:1–9.
Rossi L, Sporta Caputi S, Calizza E, Careddu G, Oliverio M, Schiaparelli S, Costantini ML 
(2019) Antarctic food web architecture under varying dynamics of sea ice cover. Sci Rep 
9:12454.
Rowley RJ (1989) Settlement and recruitment of sea urchins (Strongylocentrotus spp.) in a sea-
urchin barren ground and a kelp bed: Are populations regulated by settlement or post-
settlement processes? Mar Biol 100:485–494.
Ruess L, Müller-Navarra DC (2019) Essential biomolecules in food webs. Front Ecol Evol 7.
Rykaczewski RR, Dunne JP, Sydeman WJ, García-Reyes M, Black BA, Bograd SJ (2015) 
Poleward displacement of coastal upwelling-favorable winds in the ocean’s eastern 
boundary currents through the 21st century: Upwelling responses to climate change. 
Geophys Res Lett 42:6424–6431.
155
Sala E, Graham MH (2002) Community-wide distribution of predator–prey interaction strength 
in kelp forests. Proc Natl Acad Sci 99:3678–3683.
Salomon AK, Shears NT, Langlois TJ, Babcock RC (2008) Cascading effects of fishing can alter
carbon flow through a temperate coastal ecosystem. Ecol Appl 18:1874–1887.
Santos MAZ, Colepicolo P, Pupo D, Fujii MT, de Pereira CMP, Mesko MF (2017) Antarctic red 
macroalgae: A source of polyunsaturated fatty acids. J Appl Phycol 29:759–767.
Scheibling R, Hamm J (1991) Interactions between sea urchins (Strongylocentrotus 
droebachiensis) and their predators in field and laboratory experiments. Mar Biol 
110:105–116.
Schmid M, Kraft LGK, van der Loos LM, Kraft GT, Virtue P, Nichols PD, Hurd CL (2018) 
Southern Australian seaweeds: A promising resource for omega-3 fatty acids. Food Chem
265:70–77.
Schmitz O (2017) Predator and prey functional traits: Understanding the adaptive machinery 
driving predator–prey interactions. F1000Research 6:1767.
Schmitz OJ, Hawlena D, Trussell GC (2010) Predator control of ecosystem nutrient dynamics: 
Predator control of ecosystem nutrient dynamics. Ecol Lett 13:1199–1209.
Schneider CA, Rasband WS, Eliceiri KW (2012) NIH Image to ImageJ: 25 years of image 
analysis. Nat Methods 9:671–675.
Schram JB, Amsler MO, Galloway AWE, Amsler CD, McClintock JB (2019) Fatty acid trophic 
transfer of Antarctic algae to a sympatric amphipod consumer. Antarct Sci 31:315–316.
Schultz JA, Cloutier RN, Côté IM (2016) Evidence for a trophic cascade on rocky reefs 
following sea star mass mortality in British Columbia. PeerJ 4:e1980.
Seed R (1969) The ecology of Mytilus edulis L. (Lamellibranchiata) on exposed rocky shores. 
Oecologia 3:317–350.
Seitz RD, Wennhage H, Bergström U, Lipcius RN, Ysebaert T (2014) Ecological value of 
coastal habitats for commercially and ecologically important species. ICES J Mar Sci 
71:648–665.
Shivji M, Parker D, Hartwick B, Smith MJ (1983) Feeding and distribution study of the 
sunflower sea star Pycnopodia helianthoides (Brandt, 1835). Pac Sci 37:8.
Sloan NA, Robinson SMC (1983) Winter feeding by asteroids on a subtidal sandbed in British 
Columbia. Ophelia 22:125–140.
156
Small LF, Menzies DW (1981) Patterns of primary productivity and biomass in a coastal 
upwelling region. Deep Sea Res Part Oceanogr Res Pap 28:123–149.
Smith JG, Garcia SC (2021) Variation in purple sea urchin (Strongylocentrotus purpuratus) 
morphological traits in relation to resource availability. PeerJ 9:e11352.
Smith JG, Tomoleoni J, Staedler M, Lyon S, Fujii J, Tinker MT (2021) Behavioral responses 
across a mosaic of ecosystem states restructure a sea otter–urchin trophic cascade. Proc 
Natl Acad Sci 118:e2012493118.
Sohrabipour J (2019) Fatty acids composition of marine macroalgae (review). J Phycol Res 
3:348–374.
Spindel NB, Lee LC, Okamoto DK (2021) Metabolic depression in sea urchin barrens associated
with food deprivation. Ecology 102:e03463.
Spyksma AJP, Taylor RB, Shears NT (2017) Predation cues rather than resource availability 
promote cryptic behaviour in a habitat-forming sea urchin. Oecologia 183:821–829.
Starko S, Neufeld CJ, Gendall L, Timmer B, Campbell L, Yakimishyn J, Druehl L, Baum JK 
(2022) Microclimate predicts kelp forest extinction in the face of direct and indirect 
marine heatwave effects. Ecol Appl 32.
Stekoll MS, Deysher LE, Hess M (2006) A remote sensing approach to estimating harvestable 
kelp biomass. J Appl Phycol 18:323–334.
Steneck RS, Dethier MN (1994) A functional group approach to the structure of algal-dominated
communities. Oikos 69:476–498.
Steneck RS, Graham MH, Bourque BJ, Corbett D, Erlandson JM, Estes JA, Tegner MJ (2002) 
Kelp forest ecosystems: Biodiversity, stability, resilience and future. Environ Conserv 29.
Storero LP, Ocampo Reinaldo M, Narvarte M, Iribarne O, Botto F (2020) Trophic interactions 
and isotopic niche of octopuses and sea stars in North Patagonia. Mar Biol 167:35.
Straub SC, Wernberg T, Thomsen MS, Moore PJ, Burrows MT, Harvey BP, Smale DA (2019) 
Resistance, extinction, and everything in between – the diverse responses of seaweeds to 
marine heatwaves. Front Mar Sci 6.
Taipale SJ, Hiltunen M, Vuorio K, Peltomaa E (2016) Suitability of phytosterols alongside fatty 
acids as chemotaxonomic biomarkers for phytoplankton. Front Plant Sci 7.
157
Tait L, Bind J, Charan-Dixon H, Hawes I, Pirker J, Schiel D (2019) Unmanned aerial vehicles 
(UAVs) for monitoring macroalgal biodiversity: Comparison of RGB and multispectral 
imaging sensors for biodiversity assessments. Remote Sens 11:2332.
Tegner MJ, Dayton PK (1981) Population structure, recruitment and mortality of two sea urchins
(Strongylocentrotus franciscanus and S. purpuratus) in a kelp forest. Mar Ecol Prog Ser 
5:255–268.
Tegner MJ, Levin LA (1983) Spiny lobsters and sea urchins: Analysis of a predator-prey 
interaction. J Exp Mar Biol Ecol 73:125–150.
Teixeira TR, Santos GS, Turatti ICC, Paziani MH, von Zeska Kress MR, Colepicolo P, Debonsi 
HM (2019) Characterization of the lipid profile of Antarctic brown seaweeds and their 
endophytic fungi by gas chromatography–mass spectrometry (GC–MS). Polar Biol 
42:1431–1444.
Thompson M (2021) MaPP kelp monitoring methods.
Troch MD, Boeckx P, Cnudde C, Gansbeke DV, Vanreusel A, Vincx M, Caramujo MJ (2012) 
Bioconversion of fatty acids at the basis of marine food webs: Insights from a compound-
specific stable isotope analysis. Mar Ecol Prog Ser 465:53–67.
Vadas RL, Burrows MT, Hughes RN (1994) Foraging strategies of dogwhelks, Nucella lapillus 
(L.): Interacting effects of age, diet and chemical cues to the threat of predation. 
Oecologia 100:439–450.
Valdivia N, Aguilera M, Navarrete S, Broitman B (2015) Disentangling the effects of propagule 
supply and environmental filtering on the spatial structure of a rocky shore 
metacommunity. Mar Ecol Prog Ser 538:67–79.
Vallina SM, Ward BA, Dutkiewicz S, Follows MJ (2014) Maximal feeding with active prey-
switching: A kill-the-winner functional response and its effect on global diversity and 
biogeography. Prog Oceanogr 120:93–109.
Van Dievel M, Janssens L, Stoks R (2016) Short- and long-term behavioural, physiological and 
stoichiometric responses to predation risk indicate chronic stress and compensatory 
mechanisms. Oecologia 181:347–357.
Viana IG, Bode A (2013) Stable nitrogen isotopes in coastal macroalgae: Geographic and 
anthropogenic variability. Sci Total Environ 443:887–895.
Wen B, Gao Q, Dong S, Hou Y, Yu H (2016) Utilization of different macroalgae by sea 
cucumber Apostichopus japonicus revealed by carbon stable isotope analysis. Aquac 
Environ Interact 8:171–178.
158
Werner EE, Peacor SD (2003) A review of trait-mediated indirect interactions in ecological 
communities. Ecology 84:1083–1100.
Weykam G, Gómez I, Wiencke C, Iken K, Klöser H (1996) Photosynthetic characteristics and 
C:N ratios of macroalgae from King George Island (Antarctica). J Exp Mar Biol Ecol 
204:1–22.
Wickham H, Averick M, Bryan J, Chang W, McGowan LD, François R, Grolemund G, Hayes 
A, Henry L, Hester J, Kuhn M, Pedersen TL, Miller E, Bache SM, Müller K, Ooms J, 
Robinson D, Seidel DP, Spinu V, Takahashi K, Vaughan D, Wilke C, Woo K, Yutani H 
(2019) Welcome to the Tidyverse. J Open Source Softw 4:1686.
Wiencke C, Amsler CD (2012) Seaweeds and their communities in polar regions. In: Seaweed 
Biology: Novel Insights into Ecophysiology, Ecology and Utilization. Wiencke C, 
Bischof K (eds) Springer Berlin Heidelberg, Berlin, Heidelberg, p 265–291
Wiencke C, Amsler CD, Clayton MN (2014) Biogeographic atlas of the Southern Ocean. 
Published by The Scientific Committee on Antarctic Research, Scott Polar Reseach 
Institute, Cambridge
Wiencke C, Clayton M (2002) Antarctic seaweeds. Synopses of the Antarctic benthos. ARG 
Gantner Verl KG Ruggell Lichtenstein 9:239.
Wiencke C, Fischer G (1990) Growth and stable carbon isotope composition of cold-water 
macroalgae in relation to light and temperature. Mar Ecol Prog Ser 65:283–292.
Yu G (2022) Scatterpie: Scatter Pie Plot.
Zamzow JP, Aumack CF, Amsler CD, McClintock JB, Amsler MO, Baker BJ (2011) Gut 
contents and stable isotope analyses of the Antarctic fish, Notothenia coriiceps 
(Richardson), from two macroalgal communities. Antarct Sci 23:107–116.
Zenteno L, Cárdenas L, Valdivia N, Gómez I, Höfer J, Garrido I, Pardo L (2019) Unraveling the 
multiple bottom-up supplies of an Antarctic nearshore benthic community. Prog 
Oceanogr 174:55–63.
Zimmer R, Butman C (2000) Chemical signaling processes in the marine environment. Biol Bull 
198:168–187.
159