Southern Cross University
ePublications@SCU
Theses
2012
Trophic ecology of the nearshore zone in East
Antarctica : a stable isotope approach
Christopher L. Gillies
Southern Cross University
Publication details
Gillies, CL 2012, 'Trophic ecology of the nearshore zone in East Antarctica : a stable isotope approach', PhD thesis, Southern Cross
University, Lismore, NSW.
Copyright CL Gillies 2012
ePublications@SCU is an electronic repository administered by Southern Cross University Library. Its goal is to capture and preserve the intellectual
output of Southern Cross University authors and researchers, and to increase visibility and impact through open access to researchers around the
world. For further information please contact [email protected].
TROPHIC ECOLOGY OF THE NEARSHORE ZONE IN
EAST ANTARCTICA: A STABLE ISOTOPE APPROACH
Christopher L. Gillies BSc (Hons)
National Marine Science Centre
Southern Cross University, Coffs Harbour NSW, Australia
Australian Antarctic Division
Hobart Tasmania, Australia
Thesis submitted for the degree of Doctor of Philosophy:
Southern Cross University, February, 2012
Supervised by:
Jonathan S. Stark1, Stephen DA. Smith2 and Glenn J. Johnstone1
1
Australian Antarctic Division
Hobart TAS, Australia
2
National Marine Science Centre
Southern Cross University, Coffs Harbour NSW, Australia
“Mosquitos remind us that we are not as high up on the food chain as we think”
–Tom Wilson
Certification
I certify that the work presented in this thesis is, to the best of my
knowledge and belief, original, except as acknowledged in the text, and
that the material has not been submitted, either in whole or in part, for a
degree at this or any other university.
I acknowledge that I have read and understood the University’s rules,
requirements, procedures and policy relating to my higher degree
research award and to my thesis. I certify that I have complied with the
rules, requirements, procedures and policy of the University (as they may
be from time to time).
I warrant that I have obtained, where necessary, permission from the
copyright owners to use any third-party copyright material reproduced in
the thesis (e.g. questionnaires, artwork, unpublished letters), or to use
any of my own published work (e.g. journal articles) in which the
copyright is held by another party (e.g. publisher, co-author)
...........................................................
Christopher. L. Gillies
i
Summary
Antarctic shallow-water communities are likely to undergo considerable changes in the
near future due to the combined impacts of climate change and increasing human
occupation in coastal regions. Whilst the benthic communities of the Antarctic
Peninsula have received some attention, the shallow-water benthic communities of
Antarctica’s east coast remain relatively undescribed, hampering our ability to detect
future changes in these communities.
With this in mind, this thesis describes the food webs of two, shallow-water benthic
communities from the Windmill Islands and Vestfold Hills, East Antarctica with the
aim of establishing baseline descriptions of community structure and function. Isotopes
of carbon and nitrogen were used to delineate major energy pathways and to determine
the trophic position of consumers. This process led to the creation of a descriptive
model summarising the fundamental properties of each food web.
Both food webs shared similar features, with macroalgae and phytoplankton considered
major energy contributors, followed by benthic diatoms and to a lesser extent, sea ice
algae. There were clear energy pathways to first order-consumers, yet higher-order
consumers displayed considerable trophic diversity, resulting in complex food webs
characterised by many trophic links. Both food webs resembled trophic ‘continua’
rather than a series of discrete trophic levels, and higher-order consumers displayed
high levels of trophic omnivory.
The similarity of both food webs and to those reported from the Antarctic Peninsula, led
to a review of the global marine literature with the goals of: (1) determining if it was
possible to obtain useful food web metrics from existing published studies; and (2)
determining if these metrics could be used to detect global-scale patterns and contribute
to food web theory. The global analysis revealed that food chain length was: similar
across all habitats, averaging close to four trophic levels; weakly, though significantly,
negatively correlated with mean oceanographic primary productivity, and; positively
correlated with δ15N values of the top consumer.
ii
I consequently hypothesise that shallow-water benthic communities share similar food
web features throughout Antarctica and highlight the advantages this may have for
ecologists mapping changes to these communities brought about by human disturbance.
iii
Acknowledgements
First and foremost, I thank my supervisors: Dr Jonny Stark, Assoc. Prof. Steve Smith
and Dr Glenn Johnstone. They have shown the upmost patience, for a student who truly
needed it. I could not have asked for greater scholars, scientists and friends to work
with.
Many people from the AAD have assisted me in getting into the field; a big thank you
to all of those not specifically mentioned below. I thank the 2006/07 Casey dive team
for collecting samples prior to the commencement of this PhD and the 2008/09 Casey
sampling team for helping me collect so many samples: Cath King, Karen Miller,
Helena Baird, Glenn Johnstone and Jonny Stark. Many of those people helped again
during my second season at Davis in 2009/10: Cath King, Karen Miller, Helena Baird,
Glenn Johnstone and Jonny Stark, in addition to others: Patti Virtue, Jake van
Oosterom, Simon Reeves, Glenn Dunshea, Andrew Cawthorn, Benn Coombe and Peter
Barnes. Thank you all. Special thanks to Martin Riddle and the EP&C group, for
allowing me to be one of the team.
Debbie Lang and Trevor Bailey provided laboratory support at Kingston during my
time processing samples. Dick Williams provided advice and guidance on fish
identifications. Melissa Bastusa from Southern Cross University and Hillary StuartWilliams from the Australian National University, spent many days analysing my
samples and gave me much needed advice on stable isotopes. Ben Raymond provided
statistical advice and produced Figure 1 in Chapter VI. Kathryn James produced Figure
1 in chapters III, IV and V from maps provided by the AAD Data Centre.
A big thank you to my family and friends: who have supported me infinitely over the
last four and a half years. I couldn’t have completed this thesis without your enduring
support. Special mention to all the students in the student room and the boys in the lab
for putting up with so much of my banter! A big hug for my Antarctic buddy, Laney,
thanks so much for the endless laughs over the years.
The first half of this thesis was produced whilst I was enrolled with the University of
New England through a UNE Research Scholarship, and later in my final year, funding
and support was provided by Southern Cross University. Primary logistical and
iv
financial support for all sample collection, equipment and laboratory analysis was
provided by the Australian Antarctic Division (AAS projects 2948 and 2201) and the
National Marine Science Centre.
This research was conducted over six weeks at Casey Station in 2009 and over six
months at Davis station in 2010. I also had the opportunity to participate in a six week
Antarctic marine voyage in 2007, CEAMARC, and visited Macquarie Island for two
weeks in 2010. I attended two international conferences and obtained my Scientific and
Level 2 Commercial Divers Licence. I am truly in gratitude to all those who made this
happen, and will always cherish the memories and friendships made in such an
extraordinary place. The last four and a half years have been an absolute blast, and as
David Attenborough so eloquently calls it: it truly was ‘life in the freezer’…
…and finally, thanks to all the little creatures that make the world go round. Your lives
were sacrificed in the pursuit of good science!
v
Statement of Contribution
(& further publications)
Chapter III
Published in Polar Biology
Gillies CL, Stark JS, Smith SDA (2012) Small-scale spatial variation of δ13C and δ15N isotopes in
Antarctic carbon sources and their consumers. Polar Biol pp1-15 (in press)
I initiated the original idea, which was furthered conceptually by the second and third
authors. I collected the samples with members of the Casey 2008/09 and Davis 2009/10
field teams. I conducted all laboratory and statistical analyses and wrote the manuscript.
Manuscript drafts were edited and improved by the second and third authors and Glenn
Johnstone. Additional feedback was provided by four anonymous reviewers when
originally submitted to Marine Ecology Progress Series. Two anonymous reviewers and
the editor provided further feedback during review in Polar Biology.
Chapter IV
Published in Estuarine, Coastal and Shelf Science
Gillies CL, Stark JS, Johnstone GJ, Smith SDA (2012) Carbon flow and trophic structure of an Antarctic
coastal benthic community as determined by δ13C and δ15N. Estuar Coast Shelf Sci 97:44-57
The initial idea was created by me and the second and fourth authors. Samples were
collected by the Casey dive team in 2006/07 and by me and members of the Casey
2008/09 field team. I conducted all laboratory and statistical analyses and wrote the
manuscript. Manuscript drafts were edited and improved by the second, third and fourth
authors. Two anonymous reviewers and the editor provided useful feedback when
reviewed for Estuarine, Coastal and Shelf Science.
vi
Chapter V
A version of this manuscript has been submitted to Marine Ecology Progress Series
(Feburary 2012)
Gillies CL, Stark JS, Johnstone GJ, Smith SDA. Towards a stable isotope (δ13C, δ15N) food web model
for coastal Antarctic benthic communities: a case study from the Vestfold Hills
The initial idea was created by me and the second and fourth authors. Samples were
collected by me and the Davis field and dive teams. I conducted all laboratory and
statistical analyses and wrote the manuscript. Manuscript drafts were edited and
improved by the second, third and fourth authors.
Chapter VI
Manuscript in preparation for submission to Oikos
Gillies CL, Stark JS, and Smith SDA. A synthesis of stable isotope food webs studies from marine
systems: what can they contribute to food web theory?
The initial idea was created by me in collaboration with the second author and Glenn
Johnstone. I conducted the literature review, compiled and analysed the data and wrote
the manuscript. Ben Raymond and Steve Smith provided statistical advice. Ben
Raymond produced Figure 1 and supplied oceanic primary productivity data.
Manuscript drafts were edited and improved by the second and third authors and Glenn
Johnstone.
We warrant that the above statement of contributions is true and
accepted by all authors
Chris Gillies
Jonny Stark
Steve Smith
vii
Glenn Johnstone
Further publication relating to this thesis:
Published in Ecology
Raymond B, Marshall M, Nevitt G, Gillies CL, van den Hoff J, Stark JS, Losekoot M, Woehler EJ,
Constable AJ (2011) A Southern Ocean dietary database. Ecology 92:1188-1188.
I contributed two thirds of the stable isotope data to the database and reviewed the
manuscript prior to publication.
viii
Table of Contents
Certification ................................................................................................................................... i
Summary........................................................................................................................................ ii
Acknowledgements ....................................................................................................................... iv
Statement of Contribution ............................................................................................................ vi
I. Preface .................................................................................................................................... 12
II. Overview ............................................................................................................................... 14
Food web methodology ......................................................................................................................... 14
Stable isotopes ....................................................................................................................................... 17
The nearshore benthic ecosystem .......................................................................................................... 23
Antarctic food webs ............................................................................................................................... 28
Site descriptions ..................................................................................................................................... 29
Thesis outline ......................................................................................................................................... 33
III. Small-scale spatial variation of δ13C and δ15N isotopes in Antarctic carbon sources and
their consumers ......................................................................................................................... 35
Abstract.................................................................................................................................................. 36
Introduction ........................................................................................................................................... 37
Methods ................................................................................................................................................. 40
Results ................................................................................................................................................... 46
Discussion.............................................................................................................................................. 55
Conclusion ............................................................................................................................................. 62
Acknowledgements ............................................................................................................................... 63
References ............................................................................................................................................. 64
IV. Carbon flow and trophic structure of an Antarctic coastal benthic community as
determined by δ13C and δ15N.................................................................................................... 69
Abstract.................................................................................................................................................. 70
Introduction ........................................................................................................................................... 71
Methods ................................................................................................................................................. 74
Results ................................................................................................................................................... 79
ix
Discussion.............................................................................................................................................. 89
Conclusion ............................................................................................................................................. 99
Acknowledgements ............................................................................................................................. 100
References ........................................................................................................................................... 101
V. Food web structure of the coastal benthic community, Vestfold Hills: similarity within
East Antarctica. ....................................................................................................................... 111
Abstract................................................................................................................................................ 112
Introduction ......................................................................................................................................... 113
Methods ............................................................................................................................................... 116
Results ................................................................................................................................................. 124
Discussion............................................................................................................................................ 134
Conclusion ........................................................................................................................................... 139
Acknowledgments ............................................................................................................................... 140
References ........................................................................................................................................... 141
Appendix ............................................................................................................................................. 149
VI. A synthesis of stable isotope food webs studies from marine systems: what can they
contribute to food web theory? .............................................................................................. 151
Abstract................................................................................................................................................ 152
Introduction ......................................................................................................................................... 153
Methods ............................................................................................................................................... 156
Results ................................................................................................................................................. 160
Discussion............................................................................................................................................ 168
Conclusion ........................................................................................................................................... 175
Acknowledgements ............................................................................................................................. 176
References ........................................................................................................................................... 177
Appendix ............................................................................................................................................. 182
VII. Synthesis ........................................................................................................................... 193
Future directions .................................................................................................................................. 195
Limitations ........................................................................................................................................... 198
Concluding remarks ............................................................................................................................. 199
References (Overview and Synthesis)................................................................................................ 201
Appendicies (Published Chapters)................................................................................................... 214
x
I. Preface
Preface
During the first half of the 20th century, much of the scientific work in Antarctica was
dedicated to describing newly found species and their mechanisms of surviving in such
an extreme environment. Today, a brief look at the several dedicated Antarctic or polar
scientific journals reveals an impressive body of literature, yet studies detailing
ecosystem-wide processes are still rare in comparison to temperate or tropical systems.
Only a handful of studies have sought to determine trophic processes in the context of
whole communities (Dayton and Oliver 1977, Jarre-Teichmann et al. 1997, Kaehler et
al. 2000, Dunton 2001, Corbisier et al. 2004, Jacob et al. 2006, Norkko et al. 2007), all
of which were conducted on the Antarctic Peninsula or in the Ross Sea. The large
expanse of East Antarctica thus remains relatively unstudied. The general paucity of
food web studies is understandable, when considering the logistical and financial effort
required to operate in such an environment. Coupled with working under strict seasonal
conditions and in extreme and remote locations, it is no wonder that few large-scale
studies exist. The few studies that have been conducted were done so by countries
which have had the historical precedence, infrastructure and economic resources to
undertake large-scale scientific programs.
There is a real need to expand and build on community-wide studies in Antarctic coastal
systems. Humans are altering the environment, either directly through an increase in
human occupation and tourism, or indirectly through climate change. Both mechanisms
are hypothesised to cause changes in the distribution and abundance of species, and
ultimately, modify community structure and function. In particular, climate change will
bring wholesale changes: increasing air and water temperatures will modify sea-ice
cover, and consequently, affect light regimes and primary productivity with
repercussions for the entire food web (Massom and Stammerjohn 2010). Increased
iceberg calving from glaciers will intensify benthic scouring, increasing the loss of large
patches of benthic fauna. Ocean acidification and warmer waters will affect growth and
reproduction of many benthic species, further modifying community structure (Orr et al.
2005, Fabry et al. 2008).
Given the lack of information at scales larger than a few species, and the increasing
need for baseline studies against which to measure human-induced changes, studies
detailing carbon flow and trophic structure are clearly needed. A greater understanding
12
I. Preface
of the processes governing community composition will allow us to model, predict, and
manage the effects of climate change and human disturbance in the Antarctic coastal
marine environment. With recent advances in technology and reduced costs associated
with sample processing, ecosystem-wide studies should in the future, feature more
prominently amongst the scientific literature.
This thesis presents the first study using stable isotopes to examine carbon flow and
trophic interactions in coastal waters outside of the Antarctic Peninsula and Ross Sea.
This study provides descriptions of two shallow-water benthic communities from East
Antarctica, enabling Antarctic-wide comparisons of coastal ecosystems. By generating a
testable food web model, I am able to provide a trophic benchmark against which
modifications to these communities brought about by climate change, or other human
impacts, can be compared.
13
II. Overview
Overview
Food web methodology
Food webs are an integral component of ecology. The schematic flow of energy through
an ecosystem provides us with a theoretical canvas on which to unravel the cause and
effect of population control (Pimm 1982, Schoener 1989). In its simplest form, a food
web summarises species interactions by portraying energy-flow through a system
(Figure 1). A food web can be used to describe, hypothesise, test, determine and
compare how individuals, species, populations and communities survive. Determining
exactly which factors contribute to the survival of a species will define their
sustainability, govern population size and dictate how they interact with other species
and the environment (Pimm 1984, Leibold et al. 1997, Post 2002).
Two key data sets are needed to describe food webs: species lists, and the nature of
interactions between species. Both are difficult to obtain quantitatively and different
methodologies have the potential to produce very different outcomes. For example,
obtaining a complete list of taxa, from large mammals to tiny bacteria, for any given
location is almost impossible. Inevitably, some species will be lumped into broader
groups which reduce food chain connectivity, a key component in food web analysis
(Polis 1991, Cohen et al. 1993). In many species, different age classes have very
different feeding strategies, or diets, and it would be ideal to include each life stage into
dietary analysis for each species. This would substantially increase the number of
‘nodes’ or ‘trophic species’, providing a significant increase in food web detail,
allowing for more powerful analyses (Polis 1991). Unfortunately, collections of this
type are generally beyond the resources of most studies, and the majority of food webs
are composed using only single age classes, or ignore age classes altogether.
Determining feeding interactions is equally problematic. Traditional methods such as
visual observations and gut-content analyses are extremely time consuming and, in
smaller individuals like arthropods or crustaceans, very difficult to conduct. Traditional
techniques also suffer from their inability to provide an integrated history of feeding and
are, at best, a ‘snap-shot’ of a consumer’s diet. Whilst gut-content analysis is still
largely the best method to determine quantitative estimates of diet, its time consuming
nature, combined with the difficulty of identifying soft tissues, means it is best used in
14
II. Overview
studies that identify direct feeding relationships in single species rather than in systemwide trophic studies.
To overcome the difficulties in identifying trophic relationships and diet, ecologists
have turned to the use of indirect methods such as chemical tracer techniques. Tracer
methods, notably the use of stable isotopes and amino acid (fatty lipid) markers, are
becoming increasingly popular due to their low cost and their relative ease of
application. Both techniques involve identifying chemical signatures in tissues from an
individual which were originally derived from its diet. In essence, they mirror the
popular saying ‘you are what you eat’. Whilst neither technique currently has the ability
to provide accurate quantitative estimates of a consumers diet, they both provide an
integrated history of diet and are an ideal ‘first step’ in delineating trophic relationships.
Figure 1. Schematic diagram of the nearshore food web at Martel Inlet, King George Island. Modified
from Corbisier et al. 2004.
15
II. Overview
After undergoing a review of different methodologies, stable isotopes were selected for
use in this study due to several advantages over biochemical markers and in particular,
fatty acid analysis:
•
Stable isotope techniques are currently better developed and well known in the
scientific community and the results can be transferable across studies;
•
Stable isotope techniques are better able to provide trophic-level information in
higher-order consumers compared to fatty acid techniques;
•
Stable isotope samples are relatively simple to process and analyse, whereas
fatty acid techniques require complicated lipid extraction, which is more time
consuming than stable isotope analysis. Consequently, fewer samples can be
processed in comparative times.
Fatty acid analysis does, however, provide greater resolution when tracing carbon
sources in detrital cycles and amongst primary consumers. Where time and resources
permit, the ideal methodology would be to use a complementary approach, whereby
stable isotopes provide initial food-web structures and gut-content and fatty-acid
analyses provide more detail about individual relationships (Peterson 1999, Pasquaud et
al. 2007).
Newer methods also provide greater insight into diet analyses. Compound-specific
isotopic analysis of amino acids (Fantle et al. 1999, McClelland and Montoya 2002,
Chikaraishi et al. 2009) essentially merges both isotopic and amino acid analysis into a
single method, providing greater dietary and trophic accuracy. However, long
processing times and the need for specialist laboratories and analytical equipment,
currently preclude its use from food web studies that require a large number of samples.
DNA analysis of gut contents (e.g. Jarman et al. 2004, Dunshea 2009) are likely to be
popular in the future, especially with comprehensive genetic banking systems such as
Barcode of Life. However, they were deemed unfeasible at the start of this study due to
the large number of Antarctic species yet to be sequenced, which would limit the
interpretation of diet to higher taxonomic levels.
16
II. Overview
Stable isotopes
All elements have different isotopic forms, caused by a disproportionate number of
neutrons in the nucleus of an atom. The presence or absence of extra neutrons creates
isotopes that are either volatile, such as Uranium (238U) which decays to its counterpart
(234U), or stable, such as Hydrogen (1H, 2H), where both isotope forms are inert, safe
and persist without decay. Although all elements have stable isotopes, only a few are of
interest to ecological researchers, generally those that are tightly coupled with organic
cycles (Fry 1988). Of those, carbon isotopes (12C,
13
C) are the most well-known,
figuring strongly in the dating of soils and organic materials in antiquity studies.
However, oxygen (16O,
17
O,
18
O), sulphur (32S,
33
S,
34
S,
36
S), nitrogen (14N, 15N) and
hydrogen (1H, 2H) are also important isotopes of biological interest.
The proportions of different isotopic forms in any substance are calculated using mass
spectrometry equipment, where the ratio of the natural abundances of the heavier (e.g.
13
C,
15
N) and lighter (12C,
14
N) isotopes are compared to the abundance ratio in a
laboratory standard. The isotope composition of a sample is usually expressed in δ
notation as:
δX = [(Rsample/Rstandard-1)]*1000
where X represents the heavier isotope and R the ratio of heavy to light isotopes
(13C:12C, 15N:14N). Thus δ values are calculations of ratios of ratios and, because of the
small difference in the abundance of one isotope over another, are multiplied by 1000
for ease of use. The units of δ are per mil (‰). The most commonly used standards are:
PeeDee Belemnite (PDB) for carbon; atmospheric air for nitrogen; standard mean ocean
water (SMOW) for hydrogen and oxygen and Canyon Diablo meteorite (CD) for
sulphur. However, most standards have now been exhausted and other materials that
have been calibrated to the original standards are now used. Values for δ typically range
between -90‰ and +40‰ (Ehleringer and Rundel 1989), with negative (-) values
denoting samples which are depleted in the heavier isotope, relative to the standard.
Stable isotopes of importance to ecology generally have one isotope in far greater
abundance (Table 1).
Whilst stable isotopes are inert, the presence of extra neutrons gives each isotope a
different atomic mass, and thus isotopes display slightly different physical and chemical
properties. This allows their natural abundances to vary depending on their reactionary,
17
II. Overview
dividing or mixing forces. During kinetic reactions, a particular isotope can be
preferentially selected based on its more favourable properties, altering the natural
abundances of each isotope. The heavier isotope is generally slower to react due to
stronger bonding and, therefore, the lighter isotope is preferentially selected causing a
higher concentration of lighter isotopes relative to heavier ones (Fry 1988, Gannes et al.
1998). Alternatively, heavier isotopes can concentrate where bonds are strongest by
equilibrium (Fry 2006). Both processes result in the selection of one isotope over the
other. This process is called fractionation and when the isotope values are calculated
before and after the concentration processes, a ‘biological signature’ i.e. (fractionation
value) is obtained which is unique to the processes forming it.
The amount of fractionation is calculated by the formula:
Δ = δ SOURCE – δ PRODUCT,
where the Δ fractionation values are expressed in positive per mil units ‰. Ratios of
‘light’ to ‘heavy’ isotopes can thus be traced through organic cycles. Isotope
concentrations can also be mixed, diluting the build-up of light or heavy concentrations
back to natural abundances. Hence, isotopes flow through the environment in cycles,
eternally fractionating and mixing.
Carbon and sulphur isotopes are useful in tracing basal food sources (i.e. primary
production) in food webs as they fractionate very little from producers to consumers,
thus maintaining their isotopic signature in organism’s higher-up the food chain
(DeNiro and Epstein 1978, Fry 1988, Peterson 1999). Nitrogen is commonly used to
calculate trophic position because fractionation occurs up the food chain in a relatively
predictable stepwise fashion (Deniro and Epstein 1981, Minagawa and Wada 1984).
Oxygen and hydrogen isotopes are commonly used for tracing movements and
migration (Hobson 1999): their fractionation is a function of reactions involving H2O.
Temperature controls the evaporation of seawater and the subsequent condensation of
cloud moisture. As precipitation is correlated with latitude, altitude and temperature, the
concentration of heavy oxygen and hydrogen isotopes will differ with location (Fry
2006) and also reflect seasonal changes such as water temperature.
18
II. Overview
Table 1. Average natural abundance of stable isotopes of elements commonly used in ecological studies.
Modified from Ehleringer and Rundel (1989).
Element
Hydrogen
Carbon
Nitrogen
Oxygen
Sulphur
Isotope
Abundance (%)
1
H
99.985
2
H
0.015
12
98.89
13
C
1.11
14
N
99.63
15
N
0.37
16
O
99.759
17
O
0.037
18
O
0.204
C
32
95.00
33
0.76
34
4.22
36
0.014
S
S
S
S
By far the two most commonly used isotopes in ecological studies are carbon and
nitrogen, and a number of extensive reviews have been written on their use and
application (Rounick and Winterbourn 1986, Peterson and Fry 1987, Fry and Sherr
1989, Griffiths 1991, Wada et al. 1991, Walter et al. 1991, Gannes et al. 1998, Hobson
1999, Peterson 1999, Kelly 2000, Hood-Nowotny and Knols 2007). Tracing carbon and
nitrogen isotope abundances has proven successful across a number of key areas in
19
II. Overview
ecology. For example in: elucidating food webs, tracing energy flow or describing
feeding relationships (Kling et al. 1992, Bouillon et al. 2002, Bokhorst et al. 2007,
Carlier et al. 2007); identifying and tracking sewage flows (Costanzo et al. 2001,
Hadwen and Arthington 2007); identifying the impacts of introduced species (Rudnick
and Resh 2005); determining residency and migration (Fry et al. 2003, Hadwen et al.
2007); determining habitat preferences (Connolly 2003), and; examining ecotoxicology
(Caquet 2006). The results of isotope studies have also been instrumental in advancing
ecological theory as it relates to food webs (Vander Zanden and Fetzer 2007).
Isotopic variation in stable isotope analysis
Despite their proven usefulness, stable isotopes like any other technique, have their
caveats. Fractionation, which generates the individual biological signature necessary for
tracing organic matter, is not exactly precise and small non-dietary related variations
occur within a single individual, or among populations or species. Unaccounted
variation has the potential to bias isotopic signatures by affecting both the accuracy and
precision of stable isotope ratios, therefore creating a mismatch between an organism’s
trophic position inferred from isotope ratios, and its true trophic position. The
mechanisms for variation are well understood and, when taken into account, provide
few obstacles for precise estimates of isotopic signatures.
Primary producers provide the first step in discriminating between abundances of light
and heavy isotopes, whilst animals provide either further enrichment or dilution,
depending on the metabolic process. For producers, inter-species variation is largely
derived from different enzymes and metabolic processes associated with each
photosynthetic mode (C3, C4, CAM) (Goericke et al. 1994, Michener and Schell 1994,
Fry 2006). Differences in isotopic signatures between individuals will vary based on the
local abundance of nutrients, plant age and climatic conditions (Cloern et al. 2002).
In aquatic environments, in conjunction with those responsible for fractionation during
photosynthesis, several additional processes account for variation in carbon signatures,
resulting in larger variability across aquatic producers. Aquatic plants gain their carbon
from aqueous CO2 and/or HCO3-, which has a much higher isotopic value
(approximately 1‰) compared to aqueous CO2 (-8‰) (Peterson and Fry 1987).
Differences in their local availability or plant preference will result in different carbon
signatures. Respired CO2 is isotopically lighter than non-respired CO2 (McConnaughey
20
II. Overview
and McRoy 1979) and where autotrophs are more likely to consume respired CO2 (i.e.
stagnant and/or deeper waters), the concentration of lighter isotopes available to
producers increases, resulting in increased carbon isotope ratios (France 1995).
Additional factors contributing to variable carbon signatures include: growth rates; light
availability; and CO2-concentrating mechanisms (Wong and Sackett 1978, Rau et al.
1982, Fry and Sherr 1989, Maberly et al. 1992, Gervais and Riebesell 2001). Due to
isotopic variation, the usefulness of carbon isotopes in discriminating between primary
carbon sources (end members) is limited to those which have widely separate values
(Peterson and Fry 1987, Fry and Sherr 1989, Peterson 1999, Cloern et al. 2002).
Nitrogen variation in producers is predominantly driven by the availability and type of
nitrogen in the environment. In terrestrial systems, δ15N values can discriminate
between those plants which fix nitrogen and those that gain freely available nitrogen
from the soil (Shearer and Kohl 1989). Identifying distinct δ15N signatures at finer
scales suffers from the same variability as carbon isotopes (Cloern et al. 2002). In
addition, the δ15N values of organic matter in soils increases with depth, due to the
longer exposure of organic matter to microbial processes and are highly variable
amongst soil types (Lajtha and Marshall 1994).
Nitrogen variation in aquatic environments is predominantly driven by the availability
of nitrogen in the photic zone. In waters where nitrogen is limited, little fractionation
will occur and autotrophs will have values close to aqueous nitrogen (approx. 1‰)
(Michener and Schell 1994). In areas where nitrogen is not limited, fractionation in
primary producers is dependent on growth rates and temperature (Wada et al. 1987,
Michener and Schell 1994). Nitrogen brought into the photic zone from external
sources, such as upwelling events or terrestrial sources (e.g. sewage), will have higher
δ15N values compared to nitrogen derived from local sources (Sweeney and Kaplan
1980, Conway et al. 1994, Cabana and Rasmussen 1996).
For consumers, processes that affect the metabolic activity or biochemical make up of
an organism will contribute to isotopic variation. Variation is the result of several
causes: inherent variability; variability assimilated from producers; and variability due
to shifts in feeding regimes. Inherent variability is both inter- and intra-individual
variation, where differences between tissue types, an individual’s age, metabolic rate
and health, have the potential to produce different isotopic signatures amongst and
within individuals feeding on the same diet within similar environmental conditions
21
II. Overview
(Deniro and Epstein 1981, Hobson and Clark 1992, Gaye-Siessegger et al. 2004, GayeSiessegger et al. 2007). A consumer’s isotopic signature is derived from its diet and
therefore, variability that occurs at the level of the primary producer has the potential to
persist in higher trophic levels. The combination of inherent variability, and primary
consumer-related variability, results in uncertainty when determining changes in
isotopic signatures. Any shift in a consumer’s isotopic signature (either temporally or
spatially) cannot be attributed to a shift in diet until variation has been accounted for,
either through confirming the isotopic composition of each basal food source, or by
confirming differences in diet through gut-content analysis (Matthews and Mazumder
2004).
Whilst variation amongst producers and consumers appears overwhelming, the majority
of ‘unwanted’ variation is small and is likely to provide only minor bias in delineating
trophic positions. Furthermore, standardised laboratory and field-sampling techniques
reduce or remove much of the bias resulting from isotopic variation. We must remember
that it is isotopic variation that allows for the delineation of carbon sources and feeding
links in the first place, and several types of variation are useful for more detailed
analysis (e.g. differences in isotopic signatures between different tissue types allows for
temporal analysis of feeding regimes -Tieszen et al. 1983). By recognising the scale at
which isotopic variation takes place, and designing sampling regimes to account for
isotopic variation, we are able to mitigate most of the bias associated with isotopic
variation and provide accurate and precise isotopic signatures which match an
organism’s true trophic position.
22
II. Overview
The nearshore benthic ecosystem
The benthic fauna of the shallow (<35 m) littoral zone around the Windmill Islands
After sea ice breaks out in early summer (Nov-Dec), macroalgae rapidly increase their
biomass and cover most hard surfaces. Encrusting pink coralline algae also dominate
hard surfaces in between algal holdfasts. The red foliose algae Palmaria decipiens
dominates whilst Irdaea cordata (red) Phyllophora antarctica (red) and Monostroma
hariotii (green) are also common. In deeper waters down to depths of 50 m,
Desmarestia sp. (brown) and Himantothallus grandifolius (brown alga) are dominant.
Common fauna inhabiting algal beds include: gastropods such as Skenella
pauludionoides, spirorbid polychaetes; the isopod Cymodocella tubicauda; the
gammarid amphipod Paramoera walkeri; the ophuroid Ophiura crassa; and the fish
Trematomus bernacchii.
In areas where the sea ice is prevalent for longer periods during the summer season,
reducing or inhibiting macroalgal growth, a diverse filter-feeding invertebrate
assemblage occurs (Figure 2). On hard surfaces, sponges (Latrunculia, Homoxinella and
Isodictya species), coralline algae, hydroids, ascidians and sabellid polychaetes are
common. The grazing urchin Sterechinus neumayeri, holothurians Cucumaria sp. and
Staurocucumis sp. are also abundant. Large predators include the anemone Isotealia
antarctica, the asteroids Diplasterias brucei and Odontaster sp. and several species of
Trematomus fish.
Deeper basins within bays consist of finer sediments and are covered by a benthic mat
of diatoms that change in density and species composition throughout the year
(McMinn et al. 2004). Large rafts of unattached algae may also occupy sedimentary
basins. Common large sediment-dwelling species include the bivalves Laternula
elliptica and Adamussium colbecki and the urchin Abatus sp.. Infaunal communities are
dominated by several species of amphipod and isopod and, to a lesser extent,
cumaceans, tanaids, ostracods and polychaetes. Stark (2000) provides a detailed
description of these infaunal communities around Casey Station. The giant nemertean
Parborlasia corrugatus, and lysianassid amphipods, are common scavengers found in
all habitats.
23
II. Overview
Figure 2. Mixture of boulders and sediment with encrusting pink algae, Perkinsiana sp. polychaetes and
various sponges, Casey Station (Photo credit: Australian Antarctic Division).
Whilst there are fauna common to all types of habitats, invertebrate assemblages are
different between exposed locations on the seaward side of bays and islands, and more
sheltered areas (Johnston et al. 2007). Similarly, infaunal communities show a large
degree of variation within and between locations (Stark 2000). The heterogeneous
nature of the substrate and variation in sea ice cover are likely to account for the patchy
nature of the benthic fauna.
The benthic fauna of the shallow (<35 m) littoral zone around the Vestfold Hills
The benthic fauna occupying the shallow waters surrounding the Vestfold Hills were
surveyed during the late 1970s and early 1980s and consist of a relatively diverse
assemblage. Over 200 benthic invertebrate species occur (Tucker and Burton 1987) with
13 species of fish (Williams 1988) and 14 macroalgal species (Dhargalkar 1990). The
giant kelp Himantothallus grandifolius generally forms dense beds on boulder
substrates whilst the red algae, Iridaea cordata, Palmaria decipiens and Phyllophora
antarctica, are more common on gravel sediments. Large, unattached rafts of red
macroalgae are also common, accumulating in shallow basins. Benthic invertebrate
diversity is higher in areas where red, brown and green algae coexist (Everitt et al.
1980).
24
II. Overview
The most common invertebrate group is the Crustacea, with the amphipod Paramoera
walkeri the most numerically dominant, covering almost the entire underside of sea ice
during winter (Tucker and Burton 1988), and large patches of the substrate during
summer (pers. obs). On sandy and muddy substrates, the heart urchins Abatus sp., the
bivalve Laternula elliptica, the isopod Glyptonotus antarctica, the giant nemertean
Parborlasia corrugatus and several types of asteroids from several genera (Perkinaster,
Diplasterias, Aconodaster, Odontaster) are common. The infauna are dominated by the
amphipods Orchomenella franklini and Heterophoxus videns and the tanaid Nototanais
antarcticus. On hard surfaces (Figure 3), the urchin Sterechinus neumayeri, the
ophuroid Ophiurolepis martensi, the holothurian
Staurocucumis spatha, the
pycnogonid Nymphon australe, and the anemone Glyphoperidium bursa, are dominant
(Everitt et al. 1980, Tucker and Burton 1987, Tucker and Burton 1988). In shallow
waters (<20 m), the fish Trematomus bernacchii is numerically dominant in rocky and
weedy habitats (Williams 1988). The ice fish Chinodraco hamatus is more abundant
that T. bernacchii at depths >20 m. Trematomus newensi and Pagothenia borchgrevinki
are other common fish occurring in the region.
Figure 3. Hard surface covered in a thick mat of benthic diatoms, with the urchin Sterechinus neumayeri
and the nudibranch Notaeolidid sp. on a Perkinsiana sp. (polychaete) stem, Vestfold Hills (Photo credit: C.
Gillies).
25
II. Overview
The biota of the Windmill Islands, Vestfold Hills, and the wider Antarctic community
Many of the macroalgae, fish and invertebrate species found at the Windmill Islands are
also common to the Vestfold Hills region. Most of these species have a circum-polar
distribution and have been found to occur at Terra Nova Bay in the Ross Sea (Gambi et
al. 1994), McMurdo Sound (Dayton et al. 1974), Terre Adelie in Eastern Antarctica
(Gutt et al. 2007), and on the Antarctic Peninsula (Barnes et al. 2006). The infaunal
community at McMurdo Sound shares many of the same species found around the
Windmill Islands (Oliver and Slattery 1985), although the Windmill Islands support a
distinct crustacean-dominated assemblage (Stark 2000, Stark et al. 2003). Macroalgal
species common at the Windmill Islands are also common both on the Antarctic
Peninsula and at other sites in East Antarctica (Gambi et al. 1994, Klöser et al. 1996,
Quartino and Boraso de Zaixso 2008). The Windmill Islands supports a similar fish
assemblage to other areas in East Antarctica, such as at Dumont d’Urville (Catherine
Ozouf pers. comm.) and at Vestfold Hills (Williams 1988) although, notably, the fish
Notothenia corriceps is common on the Antarctic Peninsula and at the Windmill Islands
but is absent from the Vestfold Hills. The Windmill Islands most likely represent its
most easterly distribution (D. Williams pers. comm.).
The community structure at both the Windmill Islands and the Vestfold Hills are likely
to be similar to those described elsewhere for similar habitats (Dayton and Oliver 1977,
Gambi et al. 1994, Barnes 1995, Sahade et al. 1998, Gutt and Starmans 2003,
Echeverría and Paiva 2006). However, direct comparisons are difficult due to the
paucity of community-level studies conducted at the Windmill Islands (Stark 2000,
Johnston et al. 2007) and in the Vestfold Hills (Dhargalkar et al. 1988, Kirkwood and
Burton 1988). Apart from the fjords within the Vestfold Hills, both locations experience
no unusual physical conditions, and are subject to similar ice and light regimes.
Environmental factors influencing benthic communities
Ice is the dominant factor shaping the benthic ecosystem in shallow habitats (Barnes
1999, Gutt 2000). Anchor ice, icebergs and sea ice all provide different disturbance
regimes on the benthos and it is well documented that anchor ice and iceberg scours
cause depth zonation in shallow waters (Gambi et al. 1994, Barnes 1999, Nonato et al.
2000). Anchor ice forms on the sea floor in shallow waters during winter months,
affecting the ability of sessile fauna to colonise and occupy these areas (Dayton et al.
26
II. Overview
1969). With the additional abrasion caused by sea ice edges, areas shallower than 10 m
are depauperate, and especially devoid of sessile fauna. At greater depths, iceberg
scours can increase beta diversity by increasing heterogeneity through the presence of a
mosaic of habitats across a recovery gradient (Figure 4), comprising patches of newly
colonised, intermediate and mature assemblages (Barnes 1999, Gutt et al. 2007).
Frequent ice-scour disturbance leads to the dominance of early colonisers and lower
diversity (Dayton et al. 1974, Barnes 1995). Hence, at intermediate spatial and temporal
scales, iceberg scour provides intermediate levels of disturbance, generating niches, and
thus contributing to between-habitat diversity (Brenner et al. 2001, Knust et al. 2003).
At depths below the impact zone, assemblages are dominated by mature, slow-growing
sponge/bryozoan/ascidian communities (Barnes 1995).
The effects of sea ice on the benthic fauna are indirect but equally dramatic. Areas
covered by sea ice for long periods of time have very low levels of in situ primary
production, which can result in lower faunal densities compared to more open areas
(Dayton and Oliver 1977). These donor-controlled communities (Polis et al. 1997) rely
on advected detrital material bought in by ocean currents from more productive areas,
and thus share many attributes of deep-sea communities (Dayton and Oliver 1977).
Altered light conditions can change the density and species composition of benthic
diatom mats that cover much of the shallow soft bottom (McMinn et al. 2004), whilst a
reduction in sea-ice extent may also reduce the supply of detritus from sea ice algae
reaching the benthos. The different nutritional components of sea ice algae, compared to
phytoplankton, may alter the distributions of those species which preferentially select
ice algae for its higher nutritional content (McMahon et al. 2006). Whilst some benthic
fauna are able to change their feeding habits based on the available food source, and are
thus able to cope with the pulsed nature of productivity (Norkko et al. 2007), a
reduction in sea-ice extent, and altered ocean currents, are likely to affect trophic
relationships by altering bottom-up processes. However, the presence of benthic ‘food
banks’ (Mincks et al. 2005, Smith et al. 2006) suggests that Antarctic benthic
communities are not carbon limited: but the type of primary production, which is related
to the amount of light, plays an important role in shaping community composition.
27
II. Overview
Figure 4. The edges of an iceberg scour on the continental shelf, CEMARC voyage 2007 (Photo credit:
Australian Antarctic Division).
Antarctic food webs
Pelagic studies, and the importance of krill (Euphausia superba) as a key species,
dominate much of the work on Antarctic food webs (Siegfried et al. 1985). In the past,
Southern Ocean food webs were generalised as having short chain lengths because of
diatom-krill-macrofauna links (May 1979). However, not all of the Southern Ocean
supports high densities of krill. The Southern Ocean can be separated into three zones
based on different pelagic communities and only the seasonal pack ice zone has
densities of krill that can support large numbers of macrofauna (Hempel 1985, Hopkins
et al. 1993). The ice-free zone and permanent pack ice zone are dominated by pelagic
consumers such as salps and copepods and generally support smaller densities of larger
fauna (Hempel 1985). In all three zones, productivity is synchronised with high light
conditions in summer, after sea-ice break up initially releases sea-ice diatoms, followed
by phytoplankton blooms later in the season (Grebmeier and Barry 1991, Ducklow et al.
2006).
A large proportion of the productivity from the pelagic zone falls into the benthic zone
supporting rich suspension-feeding assemblages in areas where primary productivity is
high, and pelagic consumers have low abundances (Dayton and Oliver 1977, Clarke
1985, Ducklow et al. 2006, Smith et al. 2006). Benthic-pelagic coupling is particularly
strong in nearshore environments, where there is low biomass of pelagic consumers and
pathways to the benthos are short. Nearshore benthic systems additionally gain large
28
II. Overview
amounts of carbon from macroalgae and benthic/epilithic diatoms (Kaehler et al. 2000,
Dunton 2001, Iken et al. 2004). Several studies have shown that macroalgal fragments
are an important component of the diet of benthic consumers, providing between 2568% of ingested material (Kaehler et al. 2000, Dunton 2001). Macroalgal fragment also
support the diet of filter feeders (Tatián et al. 2008) and can be an important carbon
supplement to shelf communities (Reichardt 1987). The availability of separate carbon
sources allows benthic consumers that occupy the same trophic level to gain carbon
from different sources, increasing food web complexity (Kaehler et al. 2000, Dunton
2001).
Unlike krill in pelagic environments, there appears to be no ‘key’ species supporting
benthic food webs. Instead, higher-order consumers are supported by the rich
abundance of crustaceans, which occupy a variety of habitats and span several trophic
levels (Nyssen et al. 2002). The large diversity of prey items feeding on different carbon
sources likely results in the wide variation of δ13C values found in high-level predators
(Dunton 2001). Whilst Antarctic benthic food webs have more links than pelagic webs,
and carbon is obtained from several resources, both benthic and pelagic food webs share
similar web heights with four trophic levels being identified in both types of food webs
(Wada et al. 1987, Rau et al. 1991, Dunton 2001, Corbisier et al. 2004).
Site descriptions
The Windmill Islands
The Windmill Islands are located at 66°17' S 110°31' E on the eastern side of the
Antarctic continent on Wilkes Land, and are in close proximity to the Vanderford
Glacier (Figure 5). Spanning an area of 30 km along the coastline, the region is
characterised by 50 islands, with five main peninsulas jutting out from the Antarctic
Plateau. The peninsulas comprise low rocky hills with intervening snow-covered valleys
and meet the water’s edge with a mix of low ice edges and rocky shores (Figure 6). The
area is part of Antarctica’s small ice-free coastal zone estimated to occupy only 0.01%
of Antarctica’s coast (Snape et al. 2001). Adelie penguin colonies are common along
the rocky shorelines with a large colony located on Shirley Island adjacent to Casey
Station. The region is considered ecologically important as many other sea birds and
marine mammals use the rocky shores as breeding and haul out areas (Murray and
29
II. Overview
Figure 7. Map of Antarctica and study locations.
30
II. Overview
Luders 1990). The area has supported year-round human occupation since the
establishment of Wilkes Station in 1957 by the USA, and currently by Casey Station
(Australia). Offshore from Casey Station lies the Peterson Bank, a relatively shallow
area where icebergs (most likely calved from the Vanderford Glacier) commonly
ground.
The bathymetry at the Windmill Islands has been poorly mapped and observations have
largely been collected by divers on summer science programs. Parent bedrock consists
mainly of schists, gneisses, and migmatites (Blight and Oliver 1977). In general,
patches of bedrock and boulders occur at shallow depths near the shore, with the
substrate sloping down to finer grades of sediment. Nearshore slopes are often steep,
descending sharply to 20-30 m within 50-100 m of the shoreline, and the middle of the
bays can reach 60-100 m in depth. However, some expanses of shallow water do occur,
in particular around the Swain Group of Islands and the Browning Peninsula.
Figure 6. Looking west over Casey Station and the Windmill Islands (Photo credit: C. Gillies)
The Vestfold Hills
The Vestfold Hills lie at 68°35' S 77°58' E, 1000 km to the west of the Windmill Islands
in Prydz Bay, Prince Elizabeth Land (Figure 5). The area is bound by the Sorsdal
Glacier in the south and the Antarctic Plateau 24 km to the north, and comprises the
largest coastal ice-free area in Antarctica. Davis Station lies on a small coastal peninsula
adjacent to the shallow Heidemann Bay and is the largest of the three Australian
31
II. Overview
continental stations. The region is surrounded by low-lying rocky hills, indented by
three large fjords and several smaller bays and inlets (Figure 7). A number of small,
rocky islands, and outcrops which become submerged during high tide, lie in close
proximity to the coast. Sandy beaches and very shallow waters (<5 m) are common
across large expanses of the coastline and are utilised by large numbers of elephant seals
during the summer months. Adelie penguin colonies are common along the rocky
shorelines and islands. Halfway between Davis Station and the Sorsdal Glacier, lie the
Donskiye Island group, where several islands provide a mosaic of channels, bays and
inlets at the mouth of Ellis Fjord.
Figure 7. Collecting urchin samples in Ellis Fjord, Vestfold Hills. (Photo credit: C. Gillies)
32
II. Overview
Thesis outline
This study took place at two locations in East Antarctica: the Windmill Islands in
Wilkes Land and the Vestfold Hills in Prydz Bay, Prince Elizabeth Land (Figure 5). The
primary goal was to determine and catalogue the fundamental biological components
that constitute two shallow-water benthic ecosystems in East Antarctica. I achieve this
by defining the primary carbon sources that underpin the Windmill Islands and Vestfold
Hills food webs, and then identify the main pathways of carbon flow through the food
web to consumers.
Specifically, I use stable isotopes of carbon and nitrogen to infer diet and food web
positioning of consumers, and develop descriptive models of carbon flow from
producers to consumers, to:
1. Identify the use of different carbon sources by primary consumers;
2. Trace the flow of carbon from these sources into higher-order consumers;
3. Identify the trophic position of consumers and describe community organisation;
4. Build and test a descriptive model, which can act as a benchmark to compare
other Antarctic shallow-water food webs; and
5. Determine if comparative analysis of food webs developed from stable isotopes
can contribute to a greater understanding of food web theory.
These aims are achieved through four core publications in this thesis. I present these
chapters in manuscript format to facilitate publication in peer-reviewed journals. As
such, there may be repetition between chapters, particularly amongst the introduction
and methods sections. I use the terms ‘we’ and ‘our’ as all four data chapters contain coauthors. The Introduction and Synthesis chapters are authored solely by me except
where referenced. Reprints of published papers can be found in the Appendices of this
thesis.
Chapter III examines spatial variation in carbon and nitrogen isotopes within sediment,
algae and several consumers, in order to determine the level of replication and the scale
of sampling needed to account for naturally occurring spatial variation. It also forms the
basis of a pilot study in which I refine the field and analytic skills used in the following
33
II. Overview
studies. Chapter IV identifies carbon flow and trophic positioning of the main benthic
fauna of the Windmill Islands and generates an initial food web model. Chapter V tests
this model using an expanded dataset compiled for the Vestfold Hills and identifies
similarities between the two food webs. Chapter VI uses a global comparison of stable
isotope data, compiled from the literature, to determine the consistency of two key food
web properties- food chain length and niche width- amongst different marine systems. It
also aims to assess the state of the marine food web literature, which utilise stable
isotopes of carbon and nitrogen, to provide future studies with a better understanding of
the benefits and limitations of stable isotopes as a method of testing food web theory.
Finally, in the Synthesis I discuss how the two food webs developed are able to provide
a benchmark against which future changes to Antarctic shallow-water marine
communities can be assessed. I outline the direction for future research that should be
conducted as a natural progression from the research presented in this thesis. The
research I conducted on two expeditions, totalling nine months working in Antarctica,
will ultimately provide future researchers with the foundations on which to refine
trophic and community relationships in shallow-water benthic communities. I hope it
will provide a historical context should either community alter as a consequence of
human activities.
34
Chapter III
Chapter III
Small-scale spatial variation of δ13C and δ15N isotopes in
Antarctic carbon sources and consumers
Christopher L. Gillies, Jonathan S. Stark and Stephen DA. Smith
Published in Polar Biology
35
Chapter III
Abstract
Regional food web studies that fail to account for small-scale isotopic variability can
lead to a mismatch between an organism’s inferred and true trophic position.
Misinterpretation of trophic status may result, substantially limiting spatial and temporal
comparability of food web studies. We sampled several carbon sources and consumers
in a nested design to assess the variability of food web members across small spatial
scales (100s of m to several km) in regions around the Windmill Islands and Vestfold
Hills in East Antarctica. For carbon sources, δ13C in sea ice POM was particularly
variable between locations (km apart) and between sites (100s of m apart) with replicate
samples varying by up to 16‰. Macroalgae δ13C was less variable (replicate samples
ranging up to 6.9‰ for the red alga Iridaea cordata) yet still differed between locations.
Sediment POM and pelagic POM were the least variable, displaying minimal
differences between locations or sites for δ13C and δ15N. Three out of eight consumers
were significantly different between locations for δ13C, and five out of eight for δ15N,
with the fish Trematomus bernacchii the most variable for both δ13C and δ15N. At
smaller scales, the amphipod Paramoera walkeri showed significant variation between
sites in δ13C but not in δ15N. We attribute small-scale variability to the dynamic physical
environment for carbon sources in coastal systems and a close coupling of diet to
habitat for consumers. We highlight the need to account for small-scale spatial variation
in sampling designs for regional food web studies.
Key words: benthic food webs, East Antarctica, stable isotope analysis, trophic baseline
36
Chapter III
Introduction
The use of stable isotopes to delineate carbon sources and trophic position in food webs
is now common due to their successful application in a range of habitats, the general
availability of analytical equipment and the consequent reductions in cost. Stable
isotope methods hold an advantage over techniques such as feeding experiments and gut
content analysis because they reflect assimilated rather than potential carbon sources
and can provide a longer integrated history of feeding strategies (Hobson 1999, Fry
2006).
The basis of using isotopic measurements to delineate trophic position and carbon
source lies in documenting a regular and consistent pattern of isotopic enrichment with
increasing trophic level (Fry 1988). Nitrogen isotope ratios (14N:15N) in consumers
become enriched in δ15N by 3-4‰ with each trophic level, enabling elucidation of
trophic position (DeNiro and Epstein 1981, Minagawa and Wada 1984). Carbon ratios
(12C:13C) remain relatively stable amongst trophic levels, thus enabling carbon sources
at the base of the food web to be linked with higher-order consumers (DeNiro and
Epstein 1978, Rounick and Winterbourn 1986, Peterson and Fry 1987). Identification
and partitioning of individual carbon sources is possible when end-members have
sufficiently separate isotopic signatures (Fry and Sherr 1989, Peterson 1999).
Yet stable isotopes of carbon and nitrogen are prone to temporal and spatial variability
which, if left unaccounted, can considerably limit the interpretation of food web studies.
For example, Cabana and Rasmussen (1996) found δ15N variation of up 3.4‰ (i.e. one
trophic level) occurred amongst the same fish species within an individual lake
(variation in feeding strategies) and amongst different lakes (watershed influences on
base signature of δ15N). Simenstad et al. (1993) found variation of between 6-7‰ in
δ13C at the same site for macroalgae within a year, and as much as 3-4‰ between years.
In these cases, restricting sampling to single sites or events would result in erroneous
conclusions about carbon flow and trophic position. In many instances however, often
due to logistical or financial constraints, food web studies do collect samples from
single sites or times and present these data as a description of the regional food web.
Although it should be the aspiration of every study to sufficiently account for
variability, there is also a clear need for baseline studies to provide information on the
level of variation inherent amongst different environments and species. Such studies
37
Chapter III
provide direction in the amount of effort required to account for variability when
undertaking sampling in similar environments, and identify the level of caution that
should be factored into interpretation of isotopic results. Although there are several
studies that identify spatial or temporal variation in temperate and tropical environments
(Simenstad et al. 1993, Boon and Bunn 1994, France 1995, Cabana and Rasmussen
1996, Jennnings et al. 1997, McKinney et al 1999, Post 2002, Guest et al. 2010), only a
single study to date (Norkko et al. 2007) has directly assessed the relationships between
isotope ratios, scale and variability in Antarctic marine systems, despite their common
application in Antarctic food web studies (e.g. Kaehler et al. 2000; Dunton 2001;
Corbisier 2004; Jacob et al. 2006).
As a result of high δ15N variability in autotrophs, several studies have suggested the use
of a long-lived, primary consumer to act as a trophic baseline, as primary consumers
should be better able to integrate the temporal and spatial variation of autotrophs
(Vander Zanden and Rasmussen 1999, Post 2002). Without suitable estimates in base
values, there is no way to determine if variation in the δ15N of an organism reflects
changes in food web structure and carbon flow or just variation in the δ15N signatures
assimilated from base food sources (Vander Zanden and Rasmussen 1999, Post 2002).
To date, the identification of suitable species that can act as an isotopic baseline for
Antarctic studies has not been addressed.
Isotopic variability in autotrophs can be caused by several environmental and physicochemical factors such as light intensity, temperature and depth (Wada et al. 1987,
Michener and Schell 1994, France 1995, Hemminga and Mateo 1996). Each can affect
fractionation and assimilation of δ13C and δ15N into tissues from inorganic source
carbon and nitrogen during photosynthesis, influencing the amount of isotopic
variability (Simenstad et al. 1993, Cabana and Rasmussen 1996, Vander Zanden and
Rasmussen 1999, Aberle and Malzahn 2007). Furthermore, the origin and type of
inorganic carbon and nitrogen utilised by autotrophs can also influence isotopic
composition (Peterson and Fry 1987, Sweeney and Kaplan 1980, Conway et al. 1994,
Cabana and Rasmussen 1996). These factors operate in tandem and can occur across
several spatial scales. Therefore, failure to account for spatial differences in the δ13C
signatures of baseline autotrophs could result in an over- or underestimate of the
contribution of a base food source to a consumer’s diet (Phillips and Gregg 2001) or, for
38
Chapter III
δ15N, confound interpretation about a consumer’s trophic position (Vanden Zanden and
Rasmussen 1999, Post 2002).
Antarctic coastal waters can have strong, localised gradients in the environmental and
physico-chemical conditions that influence photosynthetic activity, and in the
availability of inorganic source carbon and nitrogen. For instance, sea ice that remains
trapped in coastal bays over the summer months can considerably decrease irradiance,
reducing macroalgal production compared to ice-free locations. Variable water retention
times and stratification due to the addition of summer melt water, combined with the
patchy addition of nutrient poor continental melt waters, may dilute the availability of
source nutrients amongst bays with different coastal aspects.
Sea ice cover and local hydrodynamic processes can also influence the spatial
distribution of food sources and their consumers in shallow waters (Dayton and Oliver
1977, Stark 2000, Johnston et al 2007, Norkko et al 2007), as can disturbance caused by
iceberg scour, resulting in highly patchy benthic communities (Gutt 2000, Smale 2008).
These conditions suggest that producers from each patch are likely to have different
isotopic signatures compared to the same producers occupying nearby patches
experiencing different physico-chemical conditions. Consumers may vary in their
isotopic signatures over small scales due to: (1) the availability of different food sources
amongst patches, and; (2) as a result of assimilating patch-specific isotopic ratios. We
therefore hypothesise that δ13C and δ15N isotope signatures in Antarctic autotrophs and
associated particulate organic matter (collectively referred to as carbon sources) will
show considerable variation across small scales of 100s of m to several km. Consumers
are also likely to show significant variation across small scales, as a result of isotopic
variability in carbon sources and food availability amongst different patches.
We measure isotopic variation amongst carbon sources and consumers across several
spatial scales and highlight the consequent implications for food web studies.
Specifically, we ask the question: Do benthic carbon sources and consumers vary in
their isotopic signature at small scales (100s of m to several km)? We also identify
several primary consumers which could act as isotopic baselines when calculating
trophic position. We sampled several carbon sources and benthic consumers using a
spatially nested design incorporating the local scale (100s of m to several km) at two
different regions: the Windmill Islands and Vestfold Hills, East Antarctica.
39
Chapter III
Methods
Survey design and sampling methods
We define small scales as those traditionally used as distances between replicated areas
within a study e.g. sites within a bay (100s of m apart) and between bays (referred to in
this study as locations, km apart). We refer to regions as larger geographic areas
encompassing several bays, an area typically defined for food web studies. Pelagic
Particulate Organic Matter (POM), sediment POM, sea ice POM 1, three algal species
and eight consumers were sampled at several locations (1000s of m to km apart) at
either the Windmill Island or Vestfold Hills region in East Antarctica (Figure 1a and
1b). Our aim was to analyse small-scale spatial patterns of stable isotopes for species
within a region but not differences between regions or species. Different regions were
sampled to increase the relevance and number of species analysed and the spatial scope
of the study. Due to the close-coupling of species to substrate (hard surfaces or soft
sediments) and sampling constraints, we were not always able to collect each species at
each location. Our analysis is therefore based on distance between groups of replicates
rather than actual differences between set locations. In addition, we were not always
able to obtain the same number of replicates at each location resulting in an unbalanced
design (Table 1). For several groups (sediment POM, sea ice POM, and Paramoera
walkeri) we were able to collect samples at spatial scales smaller than the location level.
For these groups, sites (100s of m apart) were nested within locations (1000s of m
apart). All samples within sites or locations were collected over an area with a radius of
not more than 20 m. At least three replicate samples were collected at the smallest
spatial scale, with the exceptions of pelagic POM and P. walkeri (n ≥ 2).
Sediment POM samples were collected by divers at depths of 10-30 m by coring (5 cm
diameter by 10 cm long) and utilising only the top 1.5-cm section of the core. An
homogenised sediment POM sub-sample was analysed, from which large infauna were
removed prior to drying. Pelagic POM samples were collected by horizontal tows
conducted at the surface (53 µm mesh) adjacent to the coast (approx. 100 m from the
shore) and were dominated by dinoflagellates and the diatom Trigonium antarcticum.
Sea ice POM samples were collected from fast ice adjacent to the coast by SIPRE corer
1
The term ‘Sea ice POM’ is used throughout this thesis and represents a mixture of microphytes and
bacteria. Suitable other terms include Sea Ice Microbial Community SIMCo (see Wing et al. 2011).
40
Chapter III
and comprised a mixture of diatoms and unidentified bacteria. Both pelagic POM and
sea ice POM samples were kept dark in the field at ̴ 4°C until they were brought back to
the laboratory. Samples were then examined under a dissecting microscope to remove
large zooplankton and spun in a centrifuge at 3400 rpm for 10 min: the supernatant was
discarded and the remaining sample rinsed with Milli-Q water, re-spun, dried and frozen
at -20°C for subsequent analysis.
All benthic macroalgae and consumer species (except P. walkeri and T. bernacchii)
were collected by hand at depths of 10-30 m by divers. Samples of P. walkeri were
collected using dip nets from shallow (<5 m) hard substrates. The fish T. bernacchii was
collected by long line or baited traps at depths of 8-30 m. We avoided juveniles and the
size of individuals within each species was similar. Samples were collected over a onemonth period in the summers of 2006/07 (Windmill Islands) and 2009/10 (Vestfold
Hills) and were frozen at -20°C until analysis.
41
Chapter III
Figure 1. Study area and locations (A) Vestfold Hills: (1) Flutter Is; (2) Old Wallow; (3) Readfearn Is; (4)
Davis 2; (5) Donskiye Is; (6) Magnetic Is; (7) Powell Pt; (8) Long Fjord; (9) Ice edge A; (10) Ice edge B;
(11); Ice edge C; (12) Anchorage CNL; (13) Antenna Farm; (14) Gardener Is; (15) Davis 1; (16) Warriner
Is; (17); Lake Is Reef. (B) Windmill Islands: (1) Brown Bay; (2) Honkala; (3) OB3; (4) Shannon; (5)
McGrady; (6) OB2; (7) Bailey Rks; (8) Lillianthal Is; (9) Powell Cove; (10) Shirley Is; (11) OB1; (12)
OB5; (13) Stephenson; (14) Browning; (15) Newcomb; (16) Sparkes
42
Chapter III
Stable isotope analysis
For the red macroalgae species Palmeria decipiens, Phyllophora antarctica and Iridaea
cordata, and the brown macroalgal species Desmarestia menziesii, small 1-cm2 sections
from the blade located near the stipe were removed and scraped clean of epiphytes. For
stable isotope analysis of fauna, the following tissues were used: muscle tissue from the
foot of the bivalve Laternula elliptica; the peristomial membrane for the urchin
Sterechinus neumayeri; epidermal tissue for the holothurian Cucumaria sp., ophuroid
Ophiura crassa and asteroid Diplsterias brucei; whole upper segments (1-cm section,
with feeding appendages and gut contents removed) for the sabellid polychaete
Perkinsiana sp.; muscle tissue for the fish T. bernacchii; and several whole individuals
of the amphipod P. walkeri combined (gut contents not removed).
Samples were dried in an oven at 60°C for 48 hr and ground to a fine powder.
Carbonates were removed from those samples where it was not possible to obtain
carbonate-free samples (sediment POM, S. neumayeri, O. crassa and D. brucei).
Carbonates were removed with 1 M HCL using the ‘drop-by-drop’ method
recommended by Jacob et al. (2005). Lipids were removed from samples collected at
the Windmill Islands by chemical extraction using the chloroform/methanol procedure
outlined in Logan and Lutcavage et al. (2008). Post-hoc mathematical normalisation
formulas were applied to δ13C values for samples collected from the Vestfold Hills
using the formula: ∆13C = –5.83 + 0.14 x % carbon, specified for aquatic plants and:
∆13C = –3.32 + 0.99 x C:N for aquatic animals outlined in Post et al. (2007).
Samples were analysed for δ13C and δ15N by continuous-flow isotope ratio mass
spectrometry (Thermo Delta V Plus) coupled online to a Thermo Flash 112 EA via a
Thermo Conflo III interface at the Environmental Analysis Laboratory, Southern Cross
University. The working standards used to correct for drift were Acetanilide calibrated
to NIST 8547 relative to atmospheric Nitrogen for δ15N (S.D. = 0.2 per mil) and NBS19
relative to Pee Dee Belemite for δ13C (S.D. = 0.6 per mil).
Results are expressed in the standard delta notation:
δX = ([Rsample/Rstandard-1)*1000
where X = Carbon or Nitrogen and R = the ratio of the heavy isotope over the light
isotope. Duplicates and two standards were run after every 12th sample, with the
standard deviation of duplicate material 0.09‰ for both δ13C and δ15N (n=11).
43
Chapter III
Statistical analysis
Where we had samples from locations and sites (sediment POM, sea ice POM, and P.
walkeri Table 1), we used a two-factor nested design (sites nested within locations). The
data were analysed using a two-factor univariate PERMANOVA (a permutational
analysis of variance) using a Euclidian distance similarity matrix, which produces
results equivalent to a standard ANOVA (Anderson 2001, McArdle and Anderson
2001). Variance was partitioned to calculate variance components for each level; which
relates to the magnitude of effect proportioned amongst factors, where the coefficient of
variance amongst residuals relates to the amount of unexplained variation amongst
replicates (Graham 2001). All permutations were based on raw data and all treatments
were treated as random factors. Where we had samples from different locations only
(Table 1) we used a one-factor design (location random factor). All analyses were
performed using Primer-E + PERMANOVA (v 6.1).
44
Chapter III
Table 1. Survey design and mean values of δ13C and δ15N (± SE) of carbon sources and consumers
collected in the Vestfold Hills and Windmill Islands.
Region
#Locations
#Sites
n
δ13C (%)
δ15N (%)
Pelagic POM
4
-
8
-28.9 ± 0.49
5.62 ± 0.18
Sea ice POM
8
16
47
-11.63 ± 0.70
2.95 ± 0.14
Sediment POM
3
6
23
-18.44 ± 0.23
3.76 ± 0.31
Desmarestia menziesii
2
-
15
-23.46 ± 0.43
3.91 ± 0.23
Iridaea cordata
3
-
25
-16.93 ± 0.35
3.51 ± 0.11
Palmaria decipiens
4
-
12
-18.43 ± 0.33
2.39 ± 0.18
Species
Carbon source
Vestfold Hills
Windmill Islands
Macroalgae
Consumers
Vestfold Hills
Trematomus bernacchii
7
-
126
-16.14 ± 0.08
13.98 ± 0.06
Windmill Islands
Diplasterias brucei
3
-
10
-14.77 ± 0.48
8.53 ± 0.21
Sterechinus neumayeri
3
-
21
-7.47 ± 0.22
7.43 ± 0.14
Ophiura crassa
3
-
11
-12.10 ± 0.49
8.67 ± 0.20
Paramoera walkeri
7
4
30
-13.28 ± 0.17
5.83 ± 0.11
Perkinsiana cf. antarctica
5
-
35
-16.44 ± 0.13
6.20 ± 0.07
Laternula elliptica
4
-
43
-16.63 ± 0.08
6.14 ± 0.07
Cucumaria sp.
5
-
33
-15.12 ± 0.17
8.02 ± 0.16
45
Chapter III
Results
Variation among locations
There were significant differences in δ13C and δ15N ratios amongst locations for sea ice
POM, pelagic POM, macroalgae, invertebrates and fish. Amongst carbon sources
(Table 2, Figure 2, Figure 3), pelagic POM was significantly different amongst
locations for both δ13C and δ15N, with individual samples spanning a range of 4.0‰ for
δ13C and 1.5‰ for δ15N. In sea ice POM, δ13C was significantly different amongst
locations and particularly variable amongst samples, spanning a range of 16.5‰, from 21.2 to -4.7‰, with δ15N spanning a range of 4.7‰. Sediment POM displayed no
significant difference amongst locations for δ13C and δ15N (range: 4.4‰ for δ13C, 6.0‰
for δ15N).
Amongst the macroalgae, two out of three species had significantly different δ13C
signatures between locations (P. decipiens and D. menziesii) with P. decipiens spanning
a range of 4.5‰ and D. menziesii 4.9‰. I. cordata was not significantly different
amongst locations for δ13C, although there was considerable variation amongst samples
(range 6.9‰). For δ15N (Table 2, Figure 3), only I. cordata was significantly different
between locations (range 2.4‰). D. menziesii displayed the largest variability amongst
samples for δ15N, spanning a range of 2.9‰ with P. decipiens the least variable (2.3‰).
For higher-order consumers, three out of eight species showed significant differences
amongst locations for δ13C (Table 2, Figure 4) and five out of eight species for δ15N
(Table 2, Figure 5). Two species were significantly different for both δ13C and δ15N
amongst locations: the fish T. bernacchii (spanning a range of 4.3‰ for δ13C and 3.1‰
for δ15N); and the bivalve L. elliptica (range, 2.6‰ for δ13C and 1.8‰ for δ15N).
The sabellid polychaete, Perkinsiana sp., was significantly different amongst locations
for δ13C but not for δ15N (range 3.3‰ for δ13C, 2.0‰ for δ15N). In contrast, P. walkeri,
S. neumayeri and O. crassa had significantly different isotopic values amongst locations
for δ15N but not for δ13C. In all three species, δ13C ratios spanned a range of 4-5‰. δ15N
ratios in P. walkeri and S. neumayeri spanned a range of 2.5‰, compared to 1.8‰ for
O. crassa. Cucumaria sp. and D. brucei were not significantly different in either δ13C or
δ15N amongst locations, with δ13C and δ15N samples spanning a range of 4.3‰ and
3.6‰ for Cucumaria sp. and 4.29‰ and 2.41‰ for D. brucei, respectively.
46
Chapter III
Table 2. Nested two-factor (site within location) and one-factor (location) permutational analysis of
variance (PERMANOVA) results based on Euclidean distance. Significance set at 0.05, NS = Not
Significant.
---------------δ13C---------------
---------------δ15N---------------
df
Pseudo F
Significance
Pseudo F
Significance
7,39
12.59
<0.01
2.82
NS
8,30
1.99
NS
2.51
<0.05
2,17
3.10
NS
0.80
NS
3,17
4.72
<0.05
0.62
NS
6,21
0.65
NS
27.06
<0.05
2,21
16.91
<0.01
0.19
NS
Pelagic POM
3,4
7.52
<0.05
21.17
<0.01
Palmaria decipiens
3,8
5.15
<0.05
0.88
NS
Desmarestia menziesii
3,11
4.53
<0.05
3.69
NS
Iridaea cordata
3,21
2.86
NS
4.28
<0.05
Trematomus bernacchii
6,120
19.92
<0.01
39.01
<0.01
Sterechinus neumayeri
2,18
1.58
NS
4.38
<0.05
Ophiura crassa
2,8
4.55
NS
7.54
<0.05
Cucumaria sp.
4,28
2.74
NS
2.02
NS
Diplasterias brucei
2,8
1.14
NS
0.82
NS
Laternula elliptica
3,39
3.81
<0.05
2.99
<0.05
Perkinsiana cf. antarctica
5,29
7.66
<0.01
2.00
NS
Sea ice POM Loc
Sea ice POM Site(Loc)
Sediment POM Loc
Sediment POM Site(Loc)
Paramoera walkeri Loc
Paramoera walkeri Site(Loc)
47
Chapter III
Figure 2. δ13C ratios displaying variance between locations for carbon sources. Boxes are inter-quartile
ranges, horizontal lines are medians, vertical lines are ranges with outliers (•) set at 95% confidence
intervals. Note differences in y-axis.
48
Chapter III
Figure 3. δ15N ratios displaying variance between locations for carbon sources. See Fig 2 for box-plot
descriptions. Note differences in y-axis
49
Chapter III
Figure 4. δ13C ratios displaying variance between locations for consumers. See Fig 2 for box-plot
descriptions. Note differences in y-axis.
50
Chapter III
Figure 5. δ15N ratios displaying variance between locations for consumers. See Figure 2 for box-plot
descriptions. Note differences in y-axis.
51
Chapter III
Variation among sites
There was significant variation at the site scale, for both δ13C and δ15N, and for each of
the three groups sampled (Table 2, Figure 6). δ15N for sea ice POM was significantly
different amongst sites with 23% of variance associated with the site term (Table 3).
Sediment POM δ13C values were also significantly different between sites, with 30% of
the variance associated with this term. δ13C values for P. walkeri were significantly
different between sites, with a high degree of variance (80%) associated with the site
term. In all groups, for both δ13C and δ15N, a relatively high proportion (≥ 20%) of
variation was attributed to the residual (among replicates, Table 3).
52
Chapter III
Table 3. Variance component estimates derived from nested two-factor (site within location)
permutational analysis of variance (PERMANOVA) based on Euclidean distance. Negative values treated
as zero.
-----------------δ13C----------------
Source
df
-----------------δ15N-------------------
Variance
%
Variance
%
component
Variance
component
Variance
Sea ice POM
Loc
7
16.40
70.0
0.35
35.0
Site(Loc)
8
2.37
10.1
0.23
23.1
Residual
3
4.67
19.9
0.42
41.8
Loc
2
0.57
39.7
-0.04
0.0
Site(Loc)
3
0.43
29.8
-0.25
0.0
Residual
17
0.44
30.5
2.50
100.0
Loc
6
-0.28
0.0
0.30
70.9
Site(Loc)
2
0.95
79.9
-0.02
0.0
Residual
21
0.24
20.1
0.12
29.1
Sediment POM
Paramoera
walkeri
53
Chapter III
Figure 6. δ13C and δ15N ratios displaying variance between sites (within locations) for sea ice POM,
sediment POM and P. walkeri (amphipod). See fig 2 for box-plot descriptions. Note differences in y-axis.
54
Chapter III
Discussion
This study has demonstrated that there is considerable isotopic variability in δ13C and
δ15N amongst small scales in Antarctic carbon sources and their consumers. Sea ice
POM ranged up to 6‰ for δ13C and 2‰ for δ15N between locations, whilst macroalgae
(e.g. D. menziesii up to 2.5‰ for δ13C between locations) and pelagic POM (up to 3‰
for δ13C between locations) were also variable. Three out of eight consumers for δ13C
and five out of eight consumers for δ15N also demonstrated considerable spatial
variation.
These results indicate that spatial variation is inherent in the isotopic signatures of
carbon sources and consumers, whilst demonstrating that Antarctic autotrophs and
consumers are as variable at similar spatial scales (100s of m to several km) as
temperate species. For example, over similar scales, Guest et al. (2010) found ranges in
the order of 5-10‰ for δ13C and 5‰ for δ15N in green algae in coastal regions of
Tasmania. Similarly, Simenstad et al. (1993) found δ13C variation of 6-7‰ in an annual
kelp species amongst similar sites in adjacent islands in the Bering Sea and Boon and
Bunn (1994) found δ13C variation of over 10‰ in aquatic macrophytes in billabongs in
regional Australia. Other studies have found similar levels of spatial variation amongst
consumers over similar spatial scales (France 1995, Jennings et al. 1997, McKinney et
al. 1999, Post 2002, Choy et al. 2011) highlighting the need to account for small-scale
spatial variation in regional food web studies.
Variability in carbon sources
There was consistent spatial variation amongst δ13C signatures in carbon sources, likely
reflecting spatio-temporal differences in environmental and physico-chemical
conditions (such as light intensity, temperature and depth) on isotopic fractionation
during photosynthesis (Wada et al. 1987, Michener and Schell 1994, France 1995,
Hemminga and Mateo 1996) and/or the availability and use of CO2 or HCO3- in the
environment (Simenstad et al. 1993, Cabana and Rasmussen 1996, Vander Zanden and
Rasmussen 1999, Aberle and Malzahn 2007). Local depletion of aqueous CO2 in
Antarctic coastal waters is likely over the short summer months where macroalgal
growth is high, and large beds of macroalgae occur in shallow waters, resulting in little
discrimination between
12
CO2 and
13
CO2 during photosynthetic uptake. Depleted
aqueous CO2 levels may be particularly prevalent in the sheltered, shallow bays and
55
Chapter III
islets surrounding the Vestfold Hills and Windmill Islands, where little water circulation
occurs other than tidal currents. Furthermore, δ13C signatures in producers such as
seagrass have been shown to decrease with increasing depth as a result of decreasing
irradiance (Fourqurean et al. 2007) that is also likely to occur amongst locations that
differ in the extent and duration of sea ice cover.
Similar spatio-temporal mechanisms account for nitrogen isotope variability in
autotrophs. Thus, in nitrogen-limited waters, little fractionation between source δ15N
and autotroph δ15N will occur as isotopic discrimination during photosynthesis becomes
less prevalent and autotrophs will have δ15N values close to aqueous nitrogen (approx.
1‰) (Michener and Schell 1994). In areas where nitrogen is not limited, the amount of
discrimination between source δ15N and autotrophs is dependent on growth rates and
temperature, with a general linear increase in δ15N values in producers with increasing
productivity (Wada et al. 1987, Michener and Schell 1994).
Amongst carbon sources, isotopic signatures in sea ice POM were particularly variable
amongst samples and locations, ranging over 16‰ for δ13C and 4‰ for δ15N. We found
no patterns relating to distance from shore to account for such wide variation. Samples
collected near the fast ice edge were equally as likely to have higher or lower isotopic
values (i.e. ice edge A and B versus ice edge C) as those samples collected from within
the fjord or near the coast (i.e. Long Fjord and Old Wallow sites). Small-scale
variability in sea ice POM δ15N and δ13C signatures has previously been identified at the
Vestfold Hills (Gibson et al. 1999) and in the Ross Sea (Cozzi and Cantoni 2011, Wing
et al. 2011). The exact causes of isotopic variability in sea ice POM have been linked to
temporal differences in the age of POM. Negative values are more reflective of winter
ice minima, and enriched values more reflective of the summer algal bloom, due to
reduced CO2 availability and recycling of nitrate in brine channels and the ice/water
interface during periods of high productivity (Gibson et al. 1999, Cozzi and Cantoni
2011). Variation in sea ice POM signatures may therefore primarily be the result of
seasonal differences in light, nutrients and CO2 availability; yet this manifests as spatial
variability, as under-ice erosion during the summer months creates patches of new and
old sea ice algae with differing isotopic signatures.
We found significant differences amongst locations separated by only a few kilometres
in pelagic POM. This was surprising considering that processes affecting δ13C and δ15N
signatures in phytoplankton are most likely to occur on larger spatial scales than we
56
Chapter III
sampled (e.g. Rau et al. 1982). Stratification in surface waters after sea ice breakout in
summer produces similar isotopic enriching conditions to those found in benthic
environments, whereby dissolved inorganic carbon and nitrogen become depleted due to
the lack of replenishment in stagnant waters (Gibson et al. 1999). Furthermore, the
increased surface melt water run-off from nutrient-poor continental waters into coastal
bays may reduce the concentration of marine-derived nitrogen available for uptake in
the surface waters we sampled. The different carbon signatures we found amongst
locations may therefore represent a scale of different temporal periods after sea ice
breakout and subsequent water stratification, particularly with input of fresh melt water
during summer. However, we cannot discount differences in phytoplankton
assemblages or increased input of sea ice POM amongst sites as our samples were a
mixture of phytoplankton, microzooplankton and potentially sea ice POM. Nonetheless,
these results indicate little mixing of nearshore surface waters, resulting in different
isotopic signatures for pelagic POM across the region.
We expected uniform isotopic signatures in sediment POM, as bulk POM in sediment
likely represents a time-integrated mixture of pelagic and sea ice POM in addition to
sediment microflora and bacteria. Whilst we found no differences between locations, we
did find significant differences amongst sites for δ13C and considerable variation
amongst replicates within a site for δ15N (i.e. up to 5‰, Brown Bay-site 2) indicating
possible differences in the composition of sediment POM. Small-scale physical factors
within a site, such as topography, may account for localised variation in sediment POM,
whereby small depressions may influence the ability of the substratum to capture and
retain detritus. Such small-scale processes may interact with larger-scale processes
governing the advection of sea ice algae and phytoplankton to the sea floor. Other
processes, such as sea ice cover controlling benthic light levels and consequent benthic
diatom growth (McMinn et al. 2004), could also result in non-uniform carbon
composition in surface sediments.
Some of the enriched δ15N values we found amongst producers could be the result of
localised nitrogen inputs from animal guano, derived from nearby penguin rookeries
and seal wallows. In a remote Chilean fjord, Mayr et al. (2011) postulate enriched δ15N
values in macroalgae were a result of seal guano derived from a nearby wallow, in
otherwise nitrogen-depleted waters. Although we avoided sampling directly in front of
Adelie rookeries and seal wallows at both regions (Windmill Islands and Vestfold
57
Chapter III
Hills), both areas contain high numbers of birds and seals that are active during the
summer months (Puddicombe and Johnstone 1988, Murray and Luders 1990). We
suspect similar conditions to those found by Mayr et al. (2011) may have influenced the
δ15N signatures in macroalgae to various degrees amongst different locations sampled in
our study.
Our results show the amount of variability is not equal for δ13C and δ15N, or amongst
different carbon sources, and studies utilising stable isotopes may need to closely tailor
sampling designs to account for the level of variability associated with each element or
carbon source. Applying a similar sampling effort for each carbon source would likely
result in redundant sampling effort in one carbon source, whilst failing to adequately
characterise another. Rather than utilising a ‘one size fits all’ sampling regime, we
suggest that future studies should sample carbon sources with replication based on the
amount of variability inherent in each source, as determined by pilot studies or the
literature.
Variability in consumers
Despite considerable variation amongst replicates from within a location, we were still
able to detect differences amongst populations from different locations in five out of
eight consumers for δ15N, and three out of eight for δ13C. Isotopic variability amongst
individuals from the same species (when analysing the same tissue) can be broadly
categorised into three causes: (1) variation as a result of metabolic and physical
differences between individuals (i.e. non-dietary differences); (2) variation assimilated
from original carbon sources (when diet is equal) and (3) variation resulting from
differences in prey availability amongst patches, or where individuals differ in their
prey preference (dietary differences). It is likely that, for most consumers, isotopic
variation results from a combination of all three types of variation and trophic studies
should give thought to the relative influence of each type of variation when conducting
isotopic analyses.
(1) Non-dietary sources of variation such as differences in age, sex and health can
influence fractionation rates and increase isotopic variance between individuals feeding
on the same diet (Tieszen et al. 1983, Gannes et al. 1997, Gaye-Siessegger et al. 2004).
We standardised body size to account for differences in age, yet did not account for sex.
However, it is unlikely such variability can be attributed to differences in sex alone,
58
Chapter III
particularly as the literature contains no evidence to suggest dietary differences between
sexes occurs in the species we studied. The inter-individual variation we found is most
likely to be caused by factors such as assimilation from original carbon sources (2) or
dietary differences (3).
(2) Isotopic variation amongst consumers can reflect the spatial variation in carbon
sources when the same carbon source consistently varies (spatially) at time scales
sufficient for assimilation. We found several carbon sources did vary amongst locations,
and hence, it is possible that some variation in the consumers we analysed could have
resulted from spatial variability originating from the carbon source(s) consumed.
Unfortunately, without additional temporal sampling of carbon sources to identify
temporal consistency of isotopic signatures in carbon sources, we are unable to quantify
the amount of isotopic assimilation from carbon source to consumer. However, it is
likely that at least some of the isotopic variations we found for some consumers
(particularly for sessile species) are due to variation assimilated from carbon sources
which differ in their isotopic signatures amongst patches.
For example, Lancaster and Waldron (2001) speculate that species with low mobility
and high patch affinity may have greater amongst-population variability in their isotopic
signatures and low within-population variability compared to more mobile species, as
the isotopic signature of each population is representative of the patch-specific food
source (e.g. France 1995, McKinney et al. 1999). This is likely to occur when
environmental conditions vary amongst patches but are sufficiently stable within a patch
for carbon sources to develop different isotopic signatures amongst patches. Species
that remain within a patch are likely to then obtain the patch-specific isotopic signatures
and are thus likely to vary most between patches, e.g. L. elliptica and Perkinsiana sp.
More mobile species should have lower amongst-patch isotopic variation and higher
individual variation because the isotopic signature of each individual is an average of
different patch types and more likely to be a function of individual prey preference.
However, we found no evidence of lower variation in mobile species such as T.
bernacchii, or S. neumayeri, although the lower metabolic rates found in cold water
species may increase patch affinity, due to the higher energy costs associated with wider
foraging areas.
The isotopic variability we found amongst locations may consequently reflect
differences in diet resulting from prey availability in each patch type rather than
59
Chapter III
individual prey preference (3). For example, the fish T. bernacchii displayed remarkably
different δ15N values between two sites roughly one km apart (Antenna Farm and
Gardiner Is) with a difference in mean values of 1.36‰ or close to half a trophic
position. T. bernacchii is classified as a non-discriminatory peck-and-wait predator,
matching its diet based on local prey availability (Kiest 1993, La Mesa et al. 2004). The
benthic habitat at Gardner Is is relatively deep (approx. 20 m) and composed mainly of
muddy/silt substrate with little rocky reef habitat, whilst the benthic habitat at Antenna
Farm is shallower (8-10 m) with a sandy/gravel substrate and large beds of I. cordata
and other algae. These habitats are likely to provide different prey items for T.
bernacchii resulting in the different isotopic signatures we found amongst their
populations between these locations.
Similar variation in δ15N amongst locations was found for S. neumayeri from the
Windmill Islands. The benthic communities of the Windmill Islands are largely
governed by sea ice cover and proximity to open water: exposed islands, where sea ice
breaks out early over summer, are dominated by beds of macroalgae and a larger
number of herbivores. In contrast, sheltered bays where sea ice rarely breaks out are
dominated by suspension feeders and a lower cover of macroalgae (Johnston et al.
2007). For omnivorous species such as S. neumayeri, it is likely that diet closely
matches the prey items available in each habitat. This has previously been demonstrated
by Norkko et al. (2007) at McMurdo Sound, who found different isotopic signatures in
the same species along a sea ice gradient as a function of food availability at larger
scales (10s of km). Our findings for S. neumayeri are similar to those of Norkko et al.
(2007), with small but significant differences in δ15N for different habitats (OB1sheltered, Honkala-exposed). At smaller scales, Choy et al. (2011) found considerable
variation in the diet of an intertidal grazing limpet Nacella concinna at spatial scales of
10s of m on the Antarctic Peninsula, whilst Schaal et al. (2011) found significant
differences for an Atlantic suspension feeder at the level of cm located on the top and
underside of boulders. The variable isotopic signatures amongst locations we found for
several other consumers (e.g. O. crassa, P. walkeri), in addition to T. bernacchii and S.
neumayeri, indicates that, for shallow water Antarctic communities where omnivory is a
common trait amongst many consumers, small-scale isotopic variation is likely largely a
function of prey availability.
60
Chapter III
Our results suggest that sampling designs need to account for considerable within and
amongst population spatial variability when characterising stable isotope signatures in
consumers, and species traits should be considered when interpreting causes of
variation. Consumers should be collected from the same patches as their potential food
sources to ensure their isotopic signature reflects the isotopic signatures of their
immediate prey. This is particularly important in heterogeneous environments such as
shallow waters where food availability may differ between different patches and the
isotopic signatures of carbon sources can vary across small scales as a result of the
availability of inorganic source carbon and nitrogen. In order to ensure adequate
representation of a consumer’s feeding niche through analysis of stable isotopes,
sampling regimes should consider collecting a greater number of replicates from several
different patch types.
L. elliptica and Perkinsiana sp. as isotopic baselines
As a result of high variability of δ13C and δ15N in autotrophs, several studies have
suggested the use of a long-lived primary consumer to act as a trophic baseline, as
primary consumers should be better able to integrate the temporal and spatial variation
of autotrophs (Vander Zanden and Rasmussen 1999, Post 2002). L. elliptica and
Perkinsiana sp. are both long-lived, sedentary suspension-feeding species making them
prime candidates to act as isotopic baselines. The δ15N signatures for each species were
relatively similar within a location, indicating both species may represent the patchspecific δ15N and δ13C signatures of autotrophs and hence, average smaller-scale
variation (e.g. within patch). For L. elliptica, we found δ15N also varied amongst
patches and, considering the slow metabolic and growth rate of Antarctic consumers
such as L. elliptica, any spatial variation likely represents true long-term differences in
the composition of available food or the isotopic signatures of source carbon and
nitrogen. We suggest that L. elliptica and Perkinsiana sp. should prove to be suitable
baseline species against which to define trophic position although we highlight the need
for combined dietary and isotopic studies in order to further test their effectiveness as
isotopic baselines.
61
Chapter III
Conclusion
This study, along with others (Fry and Sherr 1989, Simenstad et al. 1993, Boon and
Bunn 1994, Jennings and Warr 2003, Guest et al. 2010), indicates the need for caution
when using few samples to represent region-wide carbon flow and trophic position.
Failure to account for small-scale variability will result in potentially confounding
interpretations of the trophic positioning of consumers. In heterogeneous environments,
isotopic signatures in carbon sources and consumers are likely to vary substantially
amongst patches, at scales of 100s of m to km, based on variable physical conditions for
producers in addition to differences in the available food sources for consumers. As
Antarctic coastal benthic habitats represent particularly dynamic and patchy systems, it
is even more important to account for small-scale variability in isotopic signatures of
both carbon sources and consumers.
Although the magnitude of isotopic variability was generally less than 3‰ for δ15N
(equivalent to one trophic position) and relatively small for δ13C (with the exception of
sea ice POM), we have demonstrated that, if our study location had been treated as a
single site, then the internal site variation would have masked our ability to identify
potential localised feeding strategies amongst consumers and spatial differences in food
availability. However, we found that not all carbon sources and consumers shared a
similar level of variation and, therefore, sampling regimes should be tailored to the level
of variation associated with each carbon source and consumer species for the most
efficient use of resources. If studies wish to utilise stable isotopes of carbon and
nitrogen for regional descriptions of carbon flow and trophic structure, we suggest a
stratified sampling programme for consumers based on habitat and location, to account
for localised differences in diet amongst patches. Consumers should be collected from
the same patches as their food sources to ensure spatial variability in isotopic signatures
of carbon source does not confound trophic analysis. This would increase the precision
of dietary interpretations for regional food web studies and enable the study to identify
localised variation in feeding strategies.
62
Chapter III
Acknowledgements
We are grateful to members of the 2006/07 summer Casey field and dive team and
2009/10 summer Davis field and diving teams for assistance with field collections.
Advice on stable isotope preparations was provided by Melissa Bautista. Glenn
Johnstone provided helpful criticism on earlier versions of this Chapter. Kathryn James
produced Fig. 1 from maps generated by the Australian Antarctic Data Centre. This
research was funded by PhD research scholarships from the University of New England
and Southern Cross University and supported financially and logistically by the
Australian Antarctic Division (AAS projects 2948 and 2201).
63
Chapter III
References
Aberle N, Malzahn A (2007) Interspecific and nutrient-dependent variations in stable
isotope fractionation: experimental studies simulating pelagic multi-trophic systems.
Oecologia 154:291-303
Anderson MJ (2001) Permutation tests for univariate or multivariate analysis of
variance and regression. Can J Fish Aquat Sci 58:626-639
Boon PI, Bunn SE (1994) Variations in the stable isotope composition of aquatic plants
and their implications for food web analysis. Aquat Bot 48:99-108
Cabana G, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen
isotopes. Proc Natl Acad Sci USA 93:10844-10847
Choy EJ, Park H, Kim JH, Ahn IY, Kang CK (2011) Isotopic shift for defining habitat
exploitation by the Antarctic limpet Nacella concinna from rocky coastal habitats
(Marian Cove, King George Island). Estuar Coast Shelf Sci 92:339-346
Conway NM, Kennicutt II MC, Van Dover CL (1994) Stable isotopes in the study of
marine chemosynthetic-based ecosystems. In: Lajtha K, Michener R (eds) Stable
isotopes in ecology and environmental science. Blackwell scientific publications,
London, p 158-186
Corbisier T, Petti MV, Skowronski RP, Brito TS (2004) Trophic relationships in the
nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C stable-isotope
analysis. Polar Biol 27:75-82
Cozzi S, Cantoni C (2011) Stable isotope (δ13C and δ15N) composition of particulate
organic matter, nutrients and dissolved organic matter during spring ice retreat at Terra
Nova Bay. Antarct Sci 23:43-56
Dayton PK, Oliver JS (1977) Antarctic soft-bottom benthos in oligotrophic and
eutrophic environments. Science 197:55-58
DeNiro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in
animals. Geochim Cosmochim Acta 42:495-506
Deniro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in
animals. Geochim Cosmochim Acta 45:341-351
64
Chapter III
Dunton KH (2001) δ15N and δ13C measurements of Antarctic Peninsula fauna: trophic
relationships and assimilation of benthic seaweeds. Amer Zool 41:99-112
Fourqurean JW, Marba N, Duarte CM, Diaz-Almela E, Ruiz-Halpern S (2007) Spatial
and temporal variation in the elemental and stable isotopic content of the seagrasses
Posidonia oceanica and Cymodocea nodosa from Illes Balears, Spain. Mar Biol
151:219-232
France RL (1995) Carbon-13 enrichment in benthic compared to planktonic algae: food
web implications. Mar Ecol Prog Ser 124:307-312
Fry B (1988) Food web structure on Georges Bank from stable C, N, and S isotopic
compositions. Limnol Oceanogr 33:1182-1190
Fry B (2006) Stable isotope ecology. Springer, New York
Fry B, Sherr EB (1989)
13
C measurements as indicators of carbon flow in marine and
freshwater ecosystems. In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable isotopes
in ecological research, Vol 68. Springer-Verlag, New York, pp 196-229
Gannes LZ, O'Brien DM, del Rio CM (1997) Stable isotopes in animal ecology:
assumptions, caveats, and a call for more laboratory experiments. Ecology 78:12711276
Gaye-Siessegger J, Focken U, Muetzel S, Abel H, Becker K (2004) Feeding level and
individual metabolic rate affect δ13C and δ15N values in carp: implications for food web
studies. Oecologia 138:175-183
Gibson JAE, Trull T, Nichols PD, Summons RE, McMinn A (1999) Sedimentation of
13
C-rich organic matter from Antarctic sea-ice algae: a potential indicator of past sea-ice
extent. Geology 27:331-334
Graham MH (2001) Statistical significance versus fit: estimating the importance of
individual factors in ecological analysis of variance. Oikos 93:505-513
Guest M, Hirst A, Nichols P, Frusher S (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
Gutt J (2000) Some "driving forces" structuring communities of the sublittoral Antarctic
macrobenthos. Antarct Sci 12:297-313
65
Chapter III
Hemminga M, Mateo M (1996) Stable carbon isotopes in seagrasses: variability in
ratios and use in ecological studies. Mar Ecol Prog Ser 140:285-298
Hobson KA (1999) Tracing origins and migration of wildlife using stable isotopes: a
review. Oecologia 120:314-326
Jacob U, Brey T, Fetzer I, Kaehler S, Mintenbeck K, Dunton K, Beyer K, Struck U,
Pakhomov EA, Arntz WE (2006) Towards the trophic structure of the Bouvet Island
marine ecosystem. Polar Biol 29:106-113
Jennings S, Reñones O, Morales-Nin B, Polunin N, Moranta J, Coll J (1997) Spatial
variation in the
15
N and
13
C stable isotope composition of plants, invertebrates and
fishes on Mediterranean reefs: implications for the study of trophic pathways. Mar Ecol
Prog Ser 146:109-116
Jennings S, Warr KJ (2003) Environmental correlates of large-scale spatial variation in
the δ15N of marine animals. Mar Biol 142:1131-1140
Johnston E, Connell S, Irving A, Pile A, Gillanders B (2007) Antarctic patterns of
shallow subtidal habitat and inhabitants in Wilke’s Land. Polar Biol 30:781-788
Kaehler S, Pakhomov EA, McQuaid CD (2000) Trophic structure of the marine food
web at the Prince Edward Island (Southern Ocean) determined by 13C and 15N analysis.
Mar Ecol Prog Ser 208:13-20
Kiest KA (1993) A relationship of diet to prey abundance and the foraging behaviour of
Trematomus bernacchii. Polar Biol 13:291-296
La Mesa M, Dalú M, Vacchi M (2004) Trophic ecology of the emerald notothen
Trematomus bernacchii (pisces, nototheniidae) from Terra Nova Bay, Ross Sea,
Antarctica. Polar Biol 27:721-728
Lancaster J, Waldron S (2001) Stable isotope values of lotic invertebrates: sources of
variation, experimental design, and statistical interpretation. Limnol Oceanogr 46:723730
Logan JM, Lutcavage ME (2008) A comparison of carbon and nitrogen stable isotope
ratios of fish tissues following lipid extractions with non-polar and traditional
chloroform/methanol solvent systems. Rapid Commun Mass Sp 22:1081-1086
66
Chapter III
Mayr C, Försterra G, Häussermann V, Wunderlich A, Grau J, Zieringer M, Altenbach A
(2011) Stable isotope variability in a Chilean fjord food web: implications for n- and ccycles. Mar Ecol Prog Ser 428:89-104
McArdle BH, Anderson MJ (2001) Fitting multivariate models to community data: a
comment on distance based redundancy analysis. Ecology 82: 290-297
McKinney RA, Lake JL, Allen M, Ryba S (1999) Spatial variability in mussels used to
assess base level nitrogen isotope ratio in freshwater ecosystems. Hydrobiologia
412:17-24
Michener R, Schell DM (1994) Stable isotopes as tracers in marine aquatic food webs.
In: Lajtha K, Michener R (eds) Stable isotopes in ecology and environmental science.
Blackwell scientific publications, London, pp 138-157
Minagawa M, Wada E (1984) Stepwise enrichment of
15
N along food chains: further
evidence and the relation between δ15N and animal age. Geochim Cosmochim Acta
48:1135-1140
Murray MD, Luders DJ (1990) Faunistic studies at the Windmill Islands, Wilkes Land,
East Antarctica, 1959-80. ANARE Research Notes 73:45pp
Norkko A, Thrush SF, Cummings V, Gibbs HL, Andrew N, Norkko J, Schwarz AM
(2007) Trophic structure of coastal Antarctic food webs associated with changes in sea
ice and food supply. Ecology 88:2810-2820
Peterson BJ (1999) Stable isotopes as tracers of organic matter input and transfer in
benthic food webs: A review. Acta Oecol 20:479-487
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst
18:293-320
Phillips D, Gregg J (2001) Uncertainty in source partitioning using stable isotopes.
Oecologia 127:171-179
Post DM (2002) Using stable isotopes to estimate trophic position: models, methods,
and assumptions. Ecology 83:703-718
Post D, Layman C, Arrington D, Takimoto G, Quattrochi J, Montaña C (2007) Getting
to the fat of the matter: models, methods and assumptions for dealing with lipids in
stable isotope analyses. Oecologia 152:179-189
67
Chapter III
Puddicombe RA, Johnstone GW (1988) The breeding season diet of Adélie penguins at
the Vestfold Hills, East Antarctica. Hydrobiologia 165:239-253
Rau GH, Sweeney RE, Kaplan IR (1982) Plankton 13C:12C ratio changes with latitude:
differences between northern and southern oceans. Deep Sea Res 29:1035-1039
Rounick JS, Winterbourn MJ (1986) Stable carbon isotopes and carbon flow in
ecosystems. Bioscience 36:171-177
Schaal G, Riera P, Leroux C (2011) Microscale variations of food web functioning
within a rocky shore invertebrate community. Mar Biol 158:623-630
Simenstad CA, Duggins DO, Quay PD (1993) High turnover of inorganic carbon in
kelp habitats as a cause of δ13C variability in marine food webs. Mar Biol 116:147-160
Smale D (2008) Spatial variability in the distribution of dominant shallow-water
benthos at Adelaide Island, Antarctica. J Exp Mar Biol Ecol 357:140-148
Stark JS (2000) The distribution and abundance of soft-sediment macrobenthos around
Casey Station, East Antarctica. Polar Biol 23:840-850
Sweeney RE, Kaplan IR (1980) Tracing flocculent industrial and domestic sewage
transport on San Pedro Shelf, Southern California, by nitrogen and sulphur isotope
ratios. Mar Environ Res 3:215-224
Tieszen LL, Boutton TW, Tesdahl KG, Slade NA (1983) Fractionation and turnover of
stable carbon isotopes in animal tissues: Implications for δ13C analysis of diet.
Oecologia 57:32-37
Vander Zanden MJ, Rasmussen JB (1999) Primary consumer δ13C and δ15N and the
trophic position of aquatic consumers. Ecology 80:1395-1404
Wada E, Terazaki M, Kabaya Y, Nemoto T (1987) 15N and 13C abundances in the
Antarctic ocean with emphasis on the biogeochemical structure of the food web. DeepSea Res 34:829-841
Wing SR, Mcleod RJ, Leichter JJ, Frew RD, Lamare MD (2011) Sea ice microbial
production supports Ross Sea benthic communities: influence of a small but stable
subsidy. Ecology 93:314-323
68
Chapter IV
Chapter IV
Carbon flow and trophic structure of an Antarctic coastal
benthic community as determined by δ13C and δ15N
Christopher L. Gillies, Jonathan S. Stark, Glenn J. Johnstone and Stephen DA. Smith
Published in Estuarine, Coastal and Shelf Science
69
Chapter IV
Abstract
Stable isotopes of carbon and nitrogen were used to determine the different carbon
pathways and trophic assemblages amongst coastal benthic fauna of the Windmill
Islands, East Antarctica. Macroalgae, pelagic POM, sediment POM and sea ice POM
had well-separated δ13C signatures, which ranged from -36.75‰ for the red alga
Phyllophora antarctica, to -10.35‰ for sea ice POM. Consumers were also well
separated by δ13C, ranging from -21.42‰ for the holothurian Staurocucumis sp. up to 7.47‰ for the urchin Sterechinus neumayeri. Analysis of δ13C and δ15N revealed
distinct groups for suspension feeders, grazer/herbivores and deposit feeders, whilst
predators and predator/scavengers showed less grouping. Consumers spanned a δ15N
range of 8.71‰, equivalent to four trophic levels, although δ15N ratios amongst
consumers were continuous, rather than grouped into discrete trophic levels. We built a
trophic model for the Windmill Islands and summarised three main carbon pathways
utilised by the benthos: (1) pelagic POM; (2) macroalgae/epiphytic/benthic diatoms and
(3) sediment POM/benthic diatoms. The movement of carbon within the coastal benthic
community of the Windmill Islands is considered complex, and stable isotopes of
carbon and nitrogen were valuable tools in determining specific feeding guilds and in
tracing carbon flow, particularly amongst lower-order consumers.
Key words: stable isotope analysis, trophic ecology, Windmill Islands, Casey Station
70
Chapter IV
Introduction
Antarctic coastal ecosystems contain a diverse array of carbon sources, which support
abundant and diverse benthic communities (Dayton and Oliver 1977, Gutt et al. 2004).
During the summer months, a large proportion of sea ice algae detritus and pelagic
primary production is transferred to the benthos, creating an important coupling of
benthic communities to pelagic processes (Clarke 1985, Gibson et al. 1999, Ducklow et
al. 2006). Some of this carbon is not consumed immediately but is retained within the
sediment, providing a ‘food bank’ for suspension feeders over the winter months when
resuspended (Gili et al. 2001, Mincks et al. 2005, Smith et al. 2006). In addition to
carbon obtained from pelagic production, large beds of benthic diatoms and macroalgae
occur in shallower waters and are either consumed directly by benthic grazers (Iken
1999, Iken et al. 2004) or are assimilated through detrital pathways (Dunton 2001,
Norkko et al. 2004).
The diversity of primary production in Antarctic shallow waters is offset by strong
seasonality in the availability and palatability of carbon sources (Barnes and Clark
1995, Gibson et al. 1999, Amsler et al. 2005). Consumers have consequently developed
a variety of feeding strategies, with omnivory being particularly prevalent amongst
shallow water consumers (Arntz et al. 1994, Dayton et al. 1994). Despite diet and
feeding relationships being relatively well described for many common groups (Dayton
et al. 1974, Wägele 1989, McClintock 1994, Casaux et al. 2003, Dauby et al. 2001,
Thurber 2007 and references therein), the pathways of energy flow and trophic structure
across whole communities remains poorly understood and geographically limited. For
instance, the importance of benthic versus pelagic sources of carbon as a food source for
coastal communities has been limited to studies on only a few islands on the Antarctic
Peninsula (Kaehler et al. 2000, Dunton 2001, Corbisier et al. 2004, Jacob et al. 2006)
with only a single study at higher latitudes (Norrko et al. 2007). Furthermore, studies
detailing carbon flow in high-latitude, coastal systems are almost absent, limiting our
ability to determine how physical changes to environmental resources are likely to
affect community structure and function (Norrko et al. 2007, Massom and Stammerjohn
2010).
Antarctic food webs have previously been described as representing a ‘trophic
continuum’ (France et al. 1998), a description commonly associated with marine food
71
Chapter IV
webs (Isaacs, 1973, Polis and Strong 1996, Link 2002), rather than displaying discrete
trophic levels (Kaehler et al. 2000, Corbisier et al. 2004, Jacob et al. 2006, Mincks et al.
2008). This suggests that many Antarctic consumers display plasticity in their feeding
requirements and that Antarctic coastal food webs may represent systems that are highly
connected with high link densities between species. However, the lack of information
on carbon flows, trophic assemblages and food web linkages in Antarctic coastal
benthic systems significantly hinders the application of food web theory to Antarctic
coastal communities.
Utilising the natural abundances of carbon and nitrogen isotopes in organisms is an
efficient way of tracing carbon flow and documenting time-integrated trophic positions
in food webs. The basis of using isotopic measurements to study trophic structure lies in
documenting a regular and consistent pattern of isotopic enrichment with increasing
trophic level (Fry 1988). Nitrogen isotope ratios (14N:15N) in consumers become
enriched in δ15N by 3-4‰ with each trophic level, enabling elucidation of trophic
position (Deniro and Epstein 1981, Minagawa and Wada 1984). Carbon signatures
(12C:13C) remain relatively stable amongst trophic levels, thus enabling various carbon
sources at the base of the food web to be linked with higher-order consumers (DeNiro
and Epstein 1978, Rounick and Winterbourn 1986, Peterson and Fry 1987).
Identification and partitioning of individual carbon sources is possible when end
members have sufficiently separate isotopic signatures (Fry and Sherr 1989, Peterson
1999).
Several studies have successfully used stable isotopes to determine individual carbon
pathways and trophic relationships in Antarctic coastal benthic systems, although none
to date have been conducted in East Antarctica. Kaehler et al. (2000) were able to
determine the percentage of planktonic versus benthic carbon sources entering the
Prince Edwards Islands food web, whilst Corbisier et al. (2004) and Jacob et al. (2006)
were able to detail the close coupling of benthic consumers to pelagic production at
King George Island and Bouvet Island on the Antarctic Peninsula. Several other studies
have utilised stable isotopes of carbon and nitrogen to elucidate feeding relationships
between individual consumers and specific carbon sources (Burns et al. 1998, Dunton
2001, Nyssen et al. 2002, Norkko et al. 2004, Norkko et al. 2007).
The Windmill Islands in East Antarctica comprise 50 islands and five peninsulas within
a 30km stretch of largely ice-free coastline. Despite the area being ecologically
72
Chapter IV
important to many seabirds and megafauna (Murray and Luders 1990), little is known
of the trophic ecology of the shallow water benthic community. During the summer
months, sea ice remains trapped within bays and inlets, maintaining year round low
light levels that inhibit macroalgal growth and promote a benthic community dominated
by suspension feeding invertebrates (Stark 2000). Where sea ice breaks out over
summer, beds of macroalgae dominate, with an associated macrofaunal community of
grazing invertebrates and suspension feeders. These patterns indicate that benthic
communities at the Windmill Islands are closely coupled to the local physical
conditions controlling primary productivity.
The aim of this study was to determine the main carbon sources and to identify carbon
pathways utilised by the nearshore benthic community at the Windmill Islands.
Emphasis was placed on defining relationships between carbon sources and benthic
fauna categorised into predefined feeding guilds. We build a model utilising stable
isotopes depicting carbon flows to the benthos and identify trophic community
structure. We use the model to further explore carbon flow and trophic position in
Antarctic nearshore communities.
73
Chapter IV
Methods
Study location and sample collection 2
Casey Station is located within the Windmill Islands group at 66o17’S, 110o32’E in
Wilkes land, East Antarctica and is part of the rare coastal ice-free zone estimated to
comprise only 0.01% of Antarctica’s coastline (Snape et al. 2001) (Figure 1). Fast ice
(sea ice) in the nearshore zone is patchy during summer (December-February) creating a
mosaic of different bays which display different benthic communities (Johnston et al.
2007). In bays where sea ice breaks out early, large macroalgal beds occur in shallow
waters (1-50 m) on hard substrates, dominated by the red algae Palmaria decipiens
above 5 m depth and brown algae such as Desmarestia menziesii and Himantothallus
grandifolius below 5 m. Bays with a more sheltered aspect may remain covered in sea
ice throughout the year and have very little or no macroalgal cover. The seabed consists
of a mosaic of muddy sand, gravel, cobbles, boulders and bedrock. Samples were
collected from several sites representing a cross section of these different benthic
communities. Stark (2000) and Murray and Luders (1990) provide a more detailed
account of the benthic environment and the surrounding area.
2
Footnote: Due to the parallel nature of the field work conducted for chapters III and IV, the recommendations made
in chapter III (pg 63) which suggests future studies should alter their sampling methodology to: (1) tailor the number
of samples collected to the level of variation associated with each carbon source and consumer species; and (2)
collect samples using a stratified sampling programme to account for localised differences in diet amongst patches,
could not be strictly followed. We did however, endeavour to collect sufficient replicate material from several
locations, reducing the effect of spatial variation on our generalised descriptions of carbon flow within the Windmill
Island food web.
74
Chapter IV
Figure 1. Location of Casey Station and the Windmill Islands, East Antarctica. Stars represent sampling
sites.
Primary producers
Samples of macroalgae were hand collected by either divers on SSBA or by snorkelling.
Pelagic POM (particulate organic matter) samples were collected by horizontal tows
conducted at the surface (53 µm mesh). Epiphytic algae were scraped off subtidal rocks
(< 5 m). Samples of sea ice POM were collected from large blocks of ice debris derived
from drilling dive holes: the ice was allowed to melt in the laboratory at 4°C and filtered
onto GF/F filter papers. Consequently, samples are a mixture of sea ice diatoms,
bacteria and small metazoan heterotrophs. Pelagic POM, epiphytic algae and sea ice
POM samples were scanned under a dissecting microscope after collection to remove
larger zooplankton (copepods and ostracods), rinsed with Milli-Q water, and dried for
subsequent analysis. Sediment POM samples were collected by coring (5 cm diameter
x10 cm deep) and utilising only the top 1.5-cm section of the core. A homogenised
sediment POM subsample was analysed from which infauna were removed prior to
drying.
Consumers
Large epifaunal taxa were collected by hand by divers from depths between 5 and 30 m
and by snorkelling from depths shallower than 5 m. Infauna and small crustaceans were
75
Chapter IV
either sieved from the sediment cores or from samples collected using a small Van Veen
grab. Specimens of Paramoera walkeri were collected on hard substrates in the
shallows (<5 m) using dip nets. Nematodes were collected from within sponges at
depths between 2 and 30 m and are considered to be parasitic 3. Fish were collected by
long-line or baited traps to depths of 50 m. All samples were collected in the summers
of either 2006/07 or 2008/09 and were frozen at -20°C until analysis.
Stable isotope analysis
For larger macroalgal species (Himantothallus grandifolius, Desmarestia menziesii,
Iridaea cordata, Phyllophora antarctica) 1 cm2 sections of the blades located near the
stipe were removed and scraped clean of epiphytes. Whole fronds were scraped clean
and analysed for smaller macroalgal species (Palmaria decipiens, Monostroma sp.,
Chaetomorpha sp.). Clean white muscle tissue was used for all fish, large gastropods
and bivalves. Sections of epidermis were analysed for echinoderms (except urchins).
Whole sections (1 cm2) were analysed for sponges. The peristomial membrane was used
for the urchin Sterechinus neumayeri and intestinal tissue for the urchin Abatus sp.
Several whole individuals were combined and analysed for all small benthic crustaceans
and microgastropods. For larger polychaetes, whole upper segments (1cm section, with
feeding appendages and gut contents removed) were analysed (Perkinsiana sp.,
polynoids and nephtyids) and whole individuals for smaller worms (priapulids and
orbiniid polychaetes). All samples were thoroughly rinsed in Milli-Q water prior to
being dried in an oven at 60°C for 48 hours and ground to a fine powder.
Carbonates were removed from those samples where it was not possible to obtain
carbonate-free tissue (echinoderms, crustaceans, microgastropods, and sediment POM).
Carbonates were removed with 1M HCL using the ‘drop-by-drop’ method
recommended by Jacob et al. (2005), Kennedy et al. (2005) and Carabel et al. (2006).
Lipids were removed only from samples previously identified as having high lipid
content (mean C:N ratio >3.5 polychaetes, amphipods, sea ice POM, sediment POM
and pelagic POM (C.Gillies, unpublished data) using the chloroform/methanol
procedure outlined in Logan and Lutcavage et al. (2008).
We classified nematodes as parasitic based on the high δ15N values recorded in this study. However, it is
possible they feed on POM and DOM captured within the sponge, rather than feed on the sponge directly.
3
76
Chapter IV
Carbon (δ13C) and nitrogen (δ15N) signatures were analysed by continuous-flow isotope
ratio mass spectrometry (Thermo Delta V Plus) coupled online to a Thermo Flash 112
EA via a Thermo Conflo III interface at the Environmental Analysis Laboratory,
Southern Cross University. The working standard used for δ15N was Acetanilide
calibrated to NIST 8547 relative to atmospheric nitrogen (S.D. =0.2 per mil) and for
δ13C NBS19 relative to Pee Dee Belemite (S.D. = 0.6 per mil).
Results are expressed in the standard delta notation:
δX = ([Rsample/Rstandard-1)*1000
where X = carbon or nitrogen and R = the ratio of the heavy isotope over the light
isotope. Duplicate analysis from a variety of sample material yielded a standard
deviation of 0.19‰ for δ15N and 0.14‰ for δ13C (n=20).
Trophic categories
To facilitate analysis, we categorised each species into several trophic groups a priori
based on adult feeding mode and diet: parasite; predator; predator/scavenger; deposit
feeder; omnivore; suspension feeder and grazer/herbivore (Table 1). We define
omnivores as incorporating both plant and animal matter, and scavengers as being
necrophagous. Species were categorised based on their common feeding mode: we
acknowledge that for many species feeding categories are not definitive (for example,
many Antarctic predators or herbivores may opportunistically scavenge). Nevertheless,
trophic categories provide a suitable means of describing the typical feeding habit and
trophic status of a species, thus enabling a description of carbon flow to groups of
species with similar trophic function in the wider context of a food web.
Statistical analysis
Animals were separated by taxon and trophic category and plotted as δ13C versus δ15N
to analyse differences in diet and trophic position between different groups of animals.
To test for differences in δ13C and δ15N amongst carbon sources, we used
PERMANOVA, a permutation-based analogue of ANOVA, on a Euclidian distance
similarity matrix. PERMANOVA is more suited to unbalanced data sets and has fewer
assumptions than a standard ANOVA (Anderson 2001, McArdle and Anderson 2001).
We used the δ15N value of the scallop Adamussium colbecki as a baseline to estimate
trophic position, as long lived, primary consumers are better able to integrate the
77
Chapter IV
temporal variation associated with primary producers (Vander Zanden and Rasmussen
1999, Post 2002). We assumed a trophic fractionation of 3‰ for each trophic level
(Vander Zanden and Rasmussen 2001, Post 2002, McCutchan et al. 2003, Vanderklift
and Ponsard 2003) although we were plastic in our application based on the findings of
McCutchan et al. (2003) and Vander Zanden and Rasmussen (2001) who found lower
δ15N fractionation values in consumers raised on invertebrate diets and for aquatic
herbivores and Kaehler et al. (2000) who found slightly higher fractionation rates in
Antarctic consumers (4.1‰ for δ15N for and 1.96‰ for δ 13C).
78
Chapter IV
Results
Isotopic composition of carbon sources
Seven macroalgal species were collected including three red, two green and two brown
algae species, in addition to epiphytic algae, pelagic POM, sea ice POM and sediment
POM (Table 1). There was considerable variation in δ13C between carbon sources,
ranging from -36.75‰ for the red alga Phyllophora antarctica to -10.35‰ for sea ice
POM (Table 1, Figure 2). The red algae, I. cordata and P. decipiens had very similar
δ13C values and were grouped as one complex (Iridaea/Palmaria Complex – IPC) to
determine consumer reliance in subsequent figures (figures 3 and 4). Sediment POM
had very similar δ13C values to pelagic POM and the two most common red algae I.
cordata and P. decipiens. Results of PERMANOVA indicate significant differences
among δ13C signatures for carbon sources (F10,128= 76.63, p < 0.01), with the exception
of
the
sea
ice
POM/epiphytic
diatoms,
IPC
and
pelagic
POM/sediment
POM/Monostroma sp..
The δ15N values of carbon sources were less variable, ranging from 1.25‰ for P.
antarctica to 5.87‰ for the green alga Monostroma sp. (Table 1, Figure 2). Both green
algal species, Chaetomorpha sp. and Monostroma sp., together with epiphytic diatoms,
were more enriched in δ15N compared to the larger, more robust red and brown
macroalgal species, sea ice POM and sediment POM. Pelagic POM and P. antarctica
had similarly low δ15N values compared to the brown algae, sea ice POM and I.
cordata/P. decipiens. Results of PERMANOVA were significant (F10,128= 10.20, p
<0.01) and indicate three distinguishable groups in the δ15N plane, with Monostroma
sp./Chaetomorpha sp./epiphytic diatoms being different from
cordata/P.
decipiens/H.
grandifolius/D.
menziesii/sea
sediment POM, I.
ice
POM
and
the
P.antarctica/pelagic POM (Figure 2).
Table 1 (over page). δ13C and δ15N values (mean ± 1 SE) of organisms collected in the Windmill Islands.
PA = Parasite, P = Predator, S = Scavenger, O = Omnivore, DF = Deposit Feeder, G/H =
Grazer/Herbivore, SF = Suspension Feeder. * samples treated with lipid extraction.
79
Chapter IV
#
Trophic
category
Reference for trophic category
6.33 ± 0.15
SF
Fauchald and Jumars (1979)
-14.65 ± 0.68
10.06 ± 0.35
P
Fauchald and Jumars (1979)
23
-13.47 ± 0.16
5.61 ± 0.12
G/H
Rakusa-Suszczewski (1972)
Waldeckia obesa*
4
-18.74 ± 0.33
9.42 ± 0.11
S
Dauby et al. (2001), Nyssen et al. (2002)
Taxa
n
δ C (‰)
δ N (‰)
Epiphytic diatoms
Epiphytic diatoms
14
-11.21 ± 0.45
5.30 ± 0.45
Chlorophyla
Chaetomorpha sp.
4
-15.66 ± 0.89
5.64 ± 0.38
Monostroma sp.
11
-18.80 ± 0.67
5.92 ± 1.23
Desmarestia menziesii
12
-23.18 ± 0.46
3.76 ± 0.26
Himantothallus grandifolius
9
-20.19 ± 0.26
2.88 ± 0.32
Iridaea cordata
18
-17.38 ± 0.33
3.35 ± 0.12
Palmaria decipiens
22
-17.39 ± 0.35
3.17 ± 0.24
Phyllophora antarctica
9
-36.75 ± 0.12
1.23 ± 0.33
Sediment POM
Sediment POM*
24
-18.33 ± 0.25
3.84 ± 0.31
Sea ice POM
Sea ice POM*
17
-10.35 ± 1.56
2.85 ± 0.38
Pelagic POM
Pelagic POM*
8
-19.13 ± 0.30
1.65 ± 0.51
Perkinsiana cf. antarctica*
36
-16.30 ± 0.19
Polynoidae*
2
Paramoera walkeri*
Group
13
15
Carbon source
Phaeophyceae
Rhodophyta
Epibenthic
consumers
Annelida
1
Polychaeta
2
Arthropoda
3
Amphipoda
4
5
Isopoda
Cymodocella tubicauda
6
-13.65 ± 0.42
4.63 ± 0.38
G/H
This study
6
Pycnogonida
Nymphon australe
2
-16.55 ± 1.95
9.74 ± 0.59
P
This study
80
Chapter IV
Chordata
Notothenia coriiceps
6
-16.04 ± 0.57
13.34 ± 0.29
O
Casaux et al. (2003), Iken et al. (2004)
Trematomus bernacchii
64
-15.73 ± 0.25
12.18 ± 0.08
P/S
Kiest (1993), La Mesa et al. (2004)
Urticinopsis antarctica
15
-15.05 ± 0.35
10.61 ± 0.14
P
Dayton et al. (1970)
Psilaster charcoti
2
-10.98 ± 0.14
11.64 ± 0.24
P
McClintock (1994), This study
11
Acodontaster cf. hodgsoni
1
-11.28
11.06
P
McClintock (1994)
12
Cuenotaster involutus
1
-13.03
11.21
P
McClintock (1994)
13
Diplasterias brucei
11
-16.05 ± 0.65
8.51 ± 0.26
P/S
McClintock (1994)
14
Odontaster validus
2
-13.28 ± 0.98
10.03 ± 0.10
O/DF
McClintock (1994), Norkko et al. (2007)
Cucumaria sp.
29
-15.22 ± 0.18
8.01 ± 0.18
SF
Gutt (1991), McClintock (1994)
Staurocucumis sp.
1
-21.42
8.21
SF
Gutt (1991), McClintock (1994)
7
Perciformes
8
Cnidaria
9
Anthozoa
Echinodermata
10
15
Asteroidea
Holothuroidea
16
17
Echinoidea
Sterechinus neumayeri
21
-7.47 ± 0.22
7.43 ± 0.14
O
Pearse and Giese (1966), (McClintock (1994)
Norkko et al. (2007)
18
Ophuroidea
Ophiosparte gigas
6
-13.66 ± 0.20
9.15 ± 0.15
P
McClintock (1994), Dearborn et al. (1996)
Ophiura crassa
11
-13.74 ± 0.71
8.39 ± 0.24
DF
Lane and Riddle (2004)
Adamussium colbecki
8
-17.71 ± 0.28
5.87 ± 0.17
SF
Norkko et al. (2007)
Laternula elliptica
21
-16.58 ± 0.20
6.39 ± 0.24
SF
Ahn (1997), Norkko et al. (2007)
Neobuccinium eatoni
12
-12.83 ± 0.52
11.19 ± 0.44
P/S
Norkko et al. (2007)
Skenella paludinoides
8
-13.20 ± 0.92
5.65 ± 0.16
G/H
This study
19
Mollusca
20
Bivalvia
21
22
Gastropoda
23
24
Nematoda
Nematodes
1
-14.98
11.26
PA
This study
25
Nemertea
Parborlasia corrugatus
9
-14.69 ± 0.48
9.41 ± 0.26
S
Corbisier et al. (2004) Smale et al. (2007)
81
Chapter IV
Homaxinella sp.
4
-18.38 ± 1.20
6.22 ± 0.65
SF
McClintock et al. (2005)
27
Isodictya sp.
1
-18.96
5.87
SF
McClintock et al. (2005)
28
Unidentified sp.
4
-18.92 ± 0.51
7.21 ± 1.27
SF
McClintock et al. (2005)
Nephtyidae
5
-12.82 ± 0.79
12.01 ± 0.37
P
Fauchald and Jumars (1979)
Orbiniidae
3
-11.00 ± 0.26
8.04 ± 0.45
DF
Fauchald and Jumars (1979)
Priapulida
2
-19.99 ± 0.06
9.44 ± 0.02
P
This study
26
Porifera
Infaunal consumers
Annelida
29
Polychaeta
30
31
Priapulida
Arthropoda
32
Cumacea
Eudorella cf. splendida
2
-12.76 ± 1.37
7.35 ± 0.58
DF
Blazewicz-Paszkowycz and Ligowski (2002)
33
Ostracoda
Doloria sp.
14
-15.18 ± 1.00
6.61 ± 0.37
DF
This study
Scleroconcha sp.
17
-14.35 ± 0.59
7.58 ± 0.31
DF
This study
Heterophoxus videns
10
-13.36 ± 0.31
8.01 ± 0.62
P/S
Oliver and Slattery (1985), Dauby et al. (2001)
36
Methalimedon
nordenskjoeldi
2
-14.62 ± 0.62
9.27 ± 0.18
P/S
This study
37
Orchomenella franklini
5
-13.67 ± 1.13
5.95 ± 0.17
DF
This study
38
Orchomenella pinguides
6
-14.75 ± 0.30
8.14 ± 0.11
DF
Nyssen et al. (2002)
34
35
Amphipoda
Echinodermata
39
Echinoid
Abatus sp.
5
-10.61 ± 1.00
9.95 ± 0.69
DF
McClintock (1994), Thompson and Riddle
(2005)
40
Tanaidacea
Nototanais antarcticus
9
-14.03 ± 0.51
5.53 ± 0.39
P/DF
Oliver and Slattery (1985), BlazewiczPaszkowycz and Ligowski (2002)
Nototanais dimorphus
3
-12.56 ± 1.37
6.65 ± 1.09
P/DF
Oliver and Slattery (1985), Blazewicz-Paszkowycz and
Ligowski (2002)
41
82
Chapter IV
Figure 2. Distribution boxes of δ13C and δ15N signatures of primary producers, sediment POM, pelagic
POM and sea ice POM in the Windmill Islands (area represents ± 1 SE).
Isotopic composition of consumers
A total of 39 invertebrate and 2 fish species were collected, comprising the majority of
common benthic fauna and covering a diverse range of feeding types (Table 1). Twenty
eight species were considered epibenthic feeders, whilst 13 where considered
subsurface/deposit feeders (infauna). Consumers were well separated by δ13C, spanning
a range from -21.42‰ for the holothurian Staurocucumis sp. up to -7.47‰ for the
urchin S. neumayeri, and with the exception of S. neumayeri, were within the range of
δ13C values for carbon sources (Figure 3). The lowest δ15N values were for the grazing
isopod Cymodocella tubicauda (4.63‰ ± 0.38, mean ± SE) and the highest δ15N value
for the fish Notothenia coriiceps (13.34‰ ± 0.29), with consumers spanning a δ15N
range of 8.71‰.
Isotope signatures in suspension feeders and grazer/herbivores
Suspension feeders and grazer/herbivores formed two distinct clusters amongst the
benthos (Figure 3). The δ13C values for suspension feeders were relatively depleted,
83
Chapter IV
(group mean -16.37‰ ± 0.22) comparable to pelagic POM, with the two holothurians:
Cucumaria sp. and Staurocucumis sp., notably sitting on the extremes of the
suspension-feeding cluster. The three grazer/herbivores (P. walkeri, Skenella
pauludinoides and C. tubicauda group mean -13.64‰ ± 0.24) displayed δ13C values
comparable to epiphytic diatoms and the green alga Chaetomorpha sp. Grazer/herbivore
δ15N values (group mean 5.44‰ ± 0.11) were slightly lower than suspension feeders
(mean 6.81‰ ± 0.11), although δ15N values did not exceed 3‰ above the highest δ15N
carbon source in either group (Figure 5), consistent with signatures in first-order
consumers.
Isotope signatures in deposit feeders
Deposit feeders were loosely positioned above grazer/herbivores and were well
separated from suspension feeders in the δ13C plane (group mean -13.83‰ ± 0.30,
Figure 3). The larger surface deposit feeder Ophiura crassa (ophuroid) and smaller,
infaunal
deposit
feeders
(Eudorella
cf.
splendida,
Orchomenella
franklini,
Orchomenella pinguides, Doloria sp., Scleroconcha sp.) were closely grouped with
similar δ13C and δ15N signatures, in comparison to the relatively enriched δ13C ratios in
the heart urchin Abatus sp. and orbiniid polychaetes (Figure 4). Abatus sp. and orbiniid
polychaetes shared similar δ13C ratios to epiphytic diatoms and sea ice POM (Figures 3
and 4). With the exception of Abatus sp. deposit feeders also had δ15N values within
3‰ of carbon sources (mean 7.65‰ ± 0.16), consistent with their primary consumer
status (Figure 5).
Isotope signatures in omnivores
The three omnivorous species (S. neumayeri, N. coriiceps and Odontaster validus) had
considerably different isotopic signatures, displaying no grouping in either the δ13C or
δ15N plane (Figure 3). S. neumayeri had the most enriched δ13C values (-7.47‰ ± 0.22)
sharing a similar carbon signature to epiphytic algae. O. validus (-13.28‰ ± 0.98) had
an intermediate δ13C signature, whilst N. coriiceps (-16.04‰ ± 0.57) had the most
depleted carbon signature.
Nitrogen isotope values in N. coriiceps were the highest of any consumer (13.34‰ ±
0.29) representing the top predator in the benthic food web (Figure 5). O. validus δ15N
values were also relatively high (10.03‰ ± 0.10). S. neumayeri had the lowest δ15N
84
Chapter IV
values of the omnivores (7.43‰ ± 0.14) which were similar to the values of first-order
consumers.
Figure 3. Mean δ13C and δ15N signatures of benthic fauna of the Windmill Islands. Refer to table 1 for
species names. X = Parasite,
feeder,
= Grazer/herbivore,
= Predator,
= Predator/scavenger,
= Omnivore,
= Deposit
= Suspension Feeder. Selected primary producers are provided for
reference (δ13C mean ± SE): SI= Sea Ice POM; ED= Epiphytic Diatoms; IPC= I. cordata and P.decipiens
complex; S= Sediment POM; P= Pelagic POM; H= H.grandifolius.
85
Chapter IV
Isotope signatures in predators, predator/scavenger and parasites
Predators spanned the δ13C spectrum, although they were largely restricted to
intermediate δ13C values (group mean -14.81‰ ± 0.18, Figure 3). However several
predators had either enriched or depleted δ13C signatures representing specialised
predation on individual feeding guilds. The pycnogonid Nymphon australe (-16.55‰ ±
1.95) was considerably depleted displaying δ13C values similar to several suspension
feeders. The asteroids Psilaster charcoti (-10.98‰ ± 0.14) and Acodontaster cf.
hodgsoni (-11.28‰) were the two most enriched predators and their δ13C signatures
were similar to potential prey items: Abatus sp., S. neumayeri and O. gigas.
Both tanaid species (Nototanais dimorphus, Nototanais antarcticus) had low δ15N
values and were similar to those for infaunal deposit feeders (Figure 5). In comparison,
infaunal nephtyid polychaetes and priapulids recorded high δ15N values (12.01‰ ± 0.37
and 9.44‰ ± 0.22 respectively) corresponding to their predatory status. On the whole,
predator δ15N values were the most enriched amongst all consumers (mean 10.89‰ ±
0.12).
Predator/scavengers also spanned the range of δ13C values, signifying a diversity of
feeding regimes (group mean -15.22‰ ± 0.31). The common, large nemertean P.
corrugatus showed intermediate δ13C values, whilst the two predator/scavenger
gammarid species Methalimedon nordenskjoldi and Heterophoxus videns shared δ13C
values similar to many of the deposit feeding infauna, corresponding to their potential
prey items. The Gammarid Waldeckia obesa had the most depleted δ13C values of all
predator/scavengers (-18.74‰ ± 0.33).
Nitrogen isotope signatures amongst predator/scavengers grouped between first-order
consumers (suspension feeders, grazer/herbivores and deposit feeders) and high end
predators (Figure 5), matching their intermediary trophic status (mean 8.76‰ ± 0.18).
The intermediate δ13C value (-14.98‰) of nematodes were not similar to the δ13C
values of the sponges in which they were found (-18.75‰), however the relatively high
δ15N value (11.26‰) implies a parasitic rather than free-living habit. Unfortunately,
further diet/trophic category interpretations were restricted due to the difficulty in
obtaining replicate samples (n=1).
86
Chapter IV
Figure 4. δ13C values (mean ± SE) of fauna delineated by trophic category and (select) primary producers.
SI= Sea Ice POM, ED= Epiphytic Diatoms, IPC= I. cordata and P.decipiens complex, S= Sediment
POM, P= Pelagic POM, H= H.grandifolius.
87
Chapter IV
Figure 5. δ15N values (mean ± SE) of fauna delineated by trophic category and trophic level.
88
Chapter IV
Discussion
This study is the first to trace carbon flows and determine trophic links in a nearshore
benthic ecosystem in East Antarctica and the most comprehensive to date within high
Antarctic latitudes. Several carbon sources showed clear isotopic distinctions, which is a
key prerequisite in the use of δ15N and δ13C in deciphering the proportion of each
carbon source in a consumer’s diet, when using mixing models (Peterson 1999, Phillips
2001). Overall, carbon isotope ratios in consumers were relatively enriched compared to
pelagic POM, consistent with their benthic lifestyle (France 1995b). Isotopic signatures
of lower-order consumers showed distinct grouping along the δ13C and δ15N plane and
with the exception of deposit feeders, lower-order consumers were generally aligned
close to the δ13C values of their potential food sources. The isotopic signatures of
higher-order consumers spanned a wider array of isotopic values, which is indicative of
a range of predatory strategies and δ13C averaging in consumers with increasing trophic
level (Fry 2006). Several individual consumers had isotopic signatures decoupled from
their trophic status, indicating a different diet than previously suggested. These findings
allowed us to produce a qualitative model describing carbon flow in the Windmill
Islands, from which we were able to infer aspects of diet and trophic position amongst
common benthic fauna.
Stable isotope signatures in producers
Macroalgae in Antarctic shallow water systems are a central component of the benthic
food web (Quartino and Boraso de Zaixso 2008). They are one of the largest sources of
carbon in nearshore systems, providing an important food source for many fauna when
consumed either directly (Iken 1999, Iken et al. 1999, Iken et al. 2004) or as detritus
(Norkko et al. 2004). Macroalgae also perform a significant ecological role by creating
a structural habitat for many fauna, which increases local diversity (Dhargalkar et al.
1988, Gambi et al. 1994) and provides a surface for the more readily grazed epiphytic
algae (Thomas and Jiang 1986).
This study has demonstrated Antarctic macroalgae have well separated δ13C values, thus
facilitating the determination of feeding relationships amongst Antarctic fauna. Our
results are comparable to those for the same species elsewhere in Antarctica and reflect
the generally depleted δ13C values common to Antarctic algae. The study by Dunton
(2001) on the Antarctic Peninsula provides a comparison for several species: the red
89
Chapter IV
algae I. cordata (-19.6‰), P. decipiens (-19.9‰) and P. antarctica (-35.2‰) and the
brown alga D. menziesii (-25‰) were each within ± 2.5‰ of δ13C values recorded in
our study. The large brown alga H. grandifolius (-23.3‰) and the green alga
Monostroma sp. (-23.3‰) were slightly more variable yet still within a comparable
range (± 5‰). Norkko et al. (2004) and Norkko et al. (2007) recorded δ13C values
within ± 2‰ for P. antarctica at McMurdo Sound. Norkko et al. (2007) found similar
yet slightly more enriched values for I. cordata (-21.7 to -24.3‰). Corbisier et al.
(2004) recorded values within ± 2‰ to those recorded in this study for the brown alga
D. menziesii on King George Island on the Antarctic Peninsula.
Several studies have recorded enriched δ13C values in sea ice POM in both Arctic and
Antarctic environments when compared to pelagic phytoplankton and macroalgae
(Gibson et al. 1999, Iken et al. 2005, Søreide et al. 2006). The water/ice interface likely
provides physical conditions similar to benthic environments where a decrease in water
flow generates a stagnant boundary layer around ice algae, increasing the use of respired
CO2 and enriching δ13C values (France 1995a). Sea ice POM and epiphytic algal
signatures in our study were enriched by at least 7‰ in δ13C compared to larger
macroalgae and pelagic POM, demonstrating a clear difference between carbon sources
experiencing different physical conditions. Other benthic algae considered to experience
low turbulence, such as coralline algae and microphytobenthic algae (not sampled in
this study), are also enriched in δ13C (Dunton 2001, Corbisier et al. 2004). This provides
evidence of habitat related separation of δ13C values in producers. Larger macroalgal
species and pelagic phytoplankton that have access to greater water flow are
considerably more depleted in δ13C compared to algae that grow close to the
sediment/water and sea ice/water interface, which experience more stagnant conditions
and enriched δ13C values.
δ13C values for Pelagic POM were up to 6‰ more enriched than those found elsewhere
in shallow Antarctic waters (Kaehler et al. 2000, Corbisier et al. 2004, Norkko et al.
2007), up to 12‰ more enriched than values recorded in the Southern Ocean (Wada et
al. 1987, Nyssen et al. 2002) but similar to those found by Gibson et al. (1999) at Davis
Station in East Antarctica. Gibson et al. (1999) postulate that stratification in surface
waters during summer produces similar δ13C enriching conditions to those found in
benthic environments, whereby CO2 becomes depleted due to lack of dissolved
inorganic carbon and CO2 replenishment in stagnant waters. Our samples were collected
90
Chapter IV
in December, close to the peak phytoplankton bloom and peak melt period. The influx
of freshwater melt water into the bays surrounding the Windmill Islands would likely
produce stratification of surface waters and result in δ13C enrichment in surface
phytoplankton as a result of CO2 depletion in the surface layer. We suggest future
studies measure dissolved CO2 and salinity when collecting plankton samples to enable
a better understanding of δ13C values in plankton.
Several other causes such as differences in size fractionation (Rau et al. 1990), lipid
content (Post et al. 2007), and geographical differences in plankton biosynthesis and
metabolism (Rau et al. 1982), may also account for some of this variation. Despite
slight enrichment, our results are representative of colder waters in general, which are
more depleted in δ13C compared to temperate and tropical waters (Sackett et al. 1965),
and were within a similar range to those found in Arctic waters (Iken et al. 2005,
Tamelander et al. 2006).
We expected sediment POM values to reflect an average of the available carbon sources
and, therefore, to have an intermediate signature between the more depleted macroalgae
and pelagic POM and the more enriched sea ice POM and benthic/epiphytic diatoms.
Sediment POM values however, were more reflective of depleted macroalgae and
pelagic POM rather than enriched sea ice POM or benthic diatoms, suggesting that the
majority of sediment POM is derived from macroalgae and/or pelagic POM. Seasonal
variation in sea ice cover could also cause variability in the timing and extent of sea ice
POM reaching the sediment (Gibson et al. 1999) and therefore reduce the signature of
sea ice POM in sediment POM. In addition, where more than three carbon sources exist,
the addition of small amounts of an overly-rich or depleted carbon source (e.g. P.
antarctica) within the sediment could cause a non-proportional skew in the isotopic
signature of sediment POM (Fry 2006).
Stable isotope signatures in consumers
Suspension feeders form one of the largest components of Antarctic benthic
communities in shallow waters and on the continental shelf (Dayton et al. 1974,
McClintock et al. 2005). In general, suspension feeders of the Windmill Islands were
the most depleted in δ13C of all the consumers, indicating a reliance on the similarly
depleted pelagic POM. However, we found enriched δ13C signatures in several species
(most notably L. elliptica, P. cf. antarctica, Cucumaria sp.) consistent with the findings
91
Chapter IV
of Kaehler et al. (2000), Dunton (2001) and Tatián et al. (2008) who found that >30% of
the diet of shallow water suspension feeders were derived from macroalgal detritus. The
scallop A. colbecki did not form part of this group, despite living in close proximity to
the substrate, suggesting a diet consisting mainly of pelagic POM and the ability to
discriminate between carbon types or size fractions. The two holothurian species we
analysed had more enriched δ15N values than other suspension feeders, which is
consistent with the inclusion of zooplankton in their diet (McClintock 1994), although
depleted δ13C signatures in Staurocucumis sp. suggest other carbon sources such as
macroalgal detritus, may also be consumed.
Suspension feeders are known to resource partition components of seston, specifically
through particle size filtering of macro, nano and pico elements (Orejas et al. 2001,
Thurber 2007). At high latitudes, flagellates comprise a large proportion of diet during
winter periods when plankton biomass is low (Thurber 2007), in addition to sediment
POM through re-suspension from active suspension feeders such as L. elliptica (Ahn
1997). Feeding plasticity in many Antarctic suspension feeders has been highlighted as
a strategy combating the low pelagic production over winter months (Barnes and Clark
1995, Gili et al. 2001). Whilst isotopic analysis of carbon and nitrogen isotopes were
able to distinguish a clear planktonic signature amongst suspension feeders in this study,
our results were considerably variable and we recommend that future studies determine
the isotopic signatures of different sized plankton fractions, e.g. Rau et al. (1990), to
gain a more precises description of the different fractions utilised by suspension feeders.
The tight grouping of the three herbivores, S. paludinoides, P. walkeri and C. tubicauda
is indicative of similar feeding strategies. All three species were found living in or on
macroalgae and have been record by Dhargalkar et al. (1988) as being closely
associated with the red alga P. decipiens. P. walkeri spends the winter months grazing
on sea ice algae (Rakusa-Suszczewski, 1972, Gambi et al. 1994), however, as sea ice
POM and epiphytic algae share a similar δ13C signature we are unable to partition the
contribution of each source to their diet. Based on findings by Iken (1999) for a similar
microgastropod (Laevilacunaria antarctica), it is highly likely S. paludinoides, in
addition to P. walkeri and C. tubicauda (based on their similar isotopic signatures),
consume a mixture of macroalgae and associated epiphytic algae.
The intermediate δ13C signatures in deposit feeders strongly suggests that up to half of
their diet is derived from enriched carbon sources (sea ice POM/microphytobenthic
92
Chapter IV
diatoms) and half is derived from depleted carbon sources (pelagic POM/macroalgae
detritus). The mismatch we found between δ13C signatures in deposit feeders and
sediment POM indicates that deposit feeders may preferentially select food types from
within sediment POM. The chemical defences contained in most Antarctic macroalgae
(Amsler et al. 2005) can persist, slowing degradation and inhibiting bacterial breakdown
necessary for converting macroalgae into a detrital food source (Norkko et al. 2004).
Furthermore, deposit feeders may additionally or only assimilate/digest specific
fractions of organic matter ingested (Purinton et al. 2008). It is likely that deposit
feeders avoid, or do not metabolise, macroalgal detritus, instead selecting more of the
isotopically enriched portions of sediment POM such as sea ice detritus (McMahon et
al. 2006) or epiphytic and benthic diatoms (Blazewicz-Paszkowycz and Ligowski 2002)
in addition to bacteria and microflora. This would generate the mismatch between δ13C
values in sediment POM and deposit feeders found in this study.
Predators and predator/scavengers spanned the δ13C spectrum, with common species
such as T. bernacchii, U. antarctica, P. corrugatus and polynoid polychaetes displaying
intermediate δ13C values, indicative of their generalist feeding habits. Several predators
had either enriched or depleted δ13C signatures representing specialised predation on
particular feeding guilds (e.g. Priapulids, Psilaster charcoti, Acodontaster cf. hodgsoni).
Predators and predator/scavengers were generally enriched in δ15N compared to lowerorder consumers, consistent with their higher trophic status.
The scavenging amphipod W. obesa had relatively depleted δ13C values which were
similar to those of suspension feeders and carbon derived from pelagic sources. W.
obesa feeds predominantly on carrion (Dauby et al. 2001, Nyssen et al. 2002) although
Nyssen et al. (2002) identified small amounts of diatoms and sponges in its stomach
contents. Our results indicate W. obesa, whilst a benthic feeder, consumes energy
derived from pelagic based food sources, either through a greater consumption of
sponge material or via the consumption of carrion with original pelagic feeding habits.
As W. obesa had more intermediate δ15N values in our study, it is likely it feeds on a
greater proportion of non-carrion materials such as sponges than previously identified,
at least during the winter months when carrion may be rare.
The isotopic signatures of several species did not match the trophic category we
assigned a priori. Notably, N. coriiceps was classified as an omnivore: based on the
previous findings of Casaux et al. (2003) and Iken et al. (2004) where algae, in addition
93
Chapter IV
to benthic invertebrates, comprise a significant component of its diet. Yet the high δ15N
signature we found suggests algae contributes little to its diet and that N. coriiceps
should be considered a predator or predator/scavenger in the Windmill Islands, which is
in agreement with the recent findings of Zamzow et al. (2011) . Both Nototanais species
had δ13C and δ15N signatures more closely resembling infaunal deposit feeders, rather
than predators, which matches the findings of Blazewicz-Paszkowycz and Ligowski
(2002) yet is in contrast with the findings of Oliver and Slattery (1985), who describe
Nototanais as a predator of soft bodied infauna.
The enriched δ13C signatures of the regular urchin S. neumayeri does not match those of
its diet as identified by Pearse and Giese (1966), McClintock (1994) and Corbiser et al.
(2004) (i.e. benthic diatoms, macroalgae and occasional carrion). We are unable to
determine why S. neumayeri has such an enriched δ13C signature. It is possible that S.
neumayeri was feeding on an extremely enriched carbon source not sampled in this
study or one that has a temporally variable δ13C signature. It is also possible that the
peristomial membrane tissue analysed in this study is subject to secondary fractionation,
further enriching its δ13C signature (Tieszen et al. 1983). We removed lipids and
carbonates from S. neumayeri and found no evidence of residual components (C:N <3.5,
cessation of evanescence during carbonate acidification).
Trophic levels derived from δ15N
Based on a 3‰ enrichment in δ15N per trophic level, we identified four trophic levels
amongst the benthic fauna of the Windmill Islands, which is consistent with other
shallow water Antarctic systems (Dunton, 2001). However the linear enrichment we
found for δ15N values amongst all consumers, in addition to considerable δ15N
variability within and amongst feeding guilds, provides support for the argument that
Antarctic benthic food webs are more representative of a ‘trophic continuum’ (France et
al. 1998) rather than containing discrete trophic levels. Several other studies have
demonstrated trophic continua in Antarctica, both in shallow waters (Kaehler et al.
2000, Corbisier et al. 2004, Jacob et al. 2006) and on the continental shelf (Mincks et al.
2008).
The absence of discrete trophic levels amongst the benthic fauna is indicative of a food
web similar to the ‘multichannel’ omnivory and link complexity model outlined in Polis
and Strong (1996). Antarctic coastal benthic communities span the spectrum of feeding
94
Chapter IV
guilds and trophic levels, whilst the abundant yet spatially and temporally variable
nature of primary production dictates few primary consumers are connected solely to
one carbon source. In addition, few feeding specialists occur amongst higher-end
consumers and consequently, discreet trophic chains are lacking amongst the benthic
community. We stress, however, that several groups (most notably molluscs, isopods,
polychaetes) still lack detailed information on species-specific feeding interactions, and
few studies have sought to quantify food sources for members of lower trophic
categories.
Generalised model of carbon flow for the Windmill Islands
Limitations of stable isotope data
Determining sources of carbon through stable isotope analysis in Antarctic coastal
shallow waters suffers from many of the difficulties identified from their application to
estuarine systems (Abrantes and Sheaves 2009). Identifying isotopic signatures of
individual producers in consumers is restricted when a large number of carbon sources
are available. In addition, several carbon sources with similar isotopic signatures can
occur within close proximity to each other, making it difficult to delineate which
source(s) are consumed. Mixing models are of limited benefit when there are more food
sources than isotopes or when carbon sources have similar isotopic signatures (Phillips
2001). Hence, aspects of energy flows can only be inferred rather than quantified
through stable isotope analysis.
Distinct temporal differences are likely to occur, because isotopic signatures in
producers and consumers will vary according to season or tissue type analysed. The
majority of tissue growth in Antarctic producers occurs just prior to, or during, the short
summer season (Wiencke et al. 2007) and therefore, isotopic signatures in new tissues
or tissues with fast turnover rates will reflect recent diet, whislt older tissues may be
more reflective of winter or previous summers diet. The tissue type analysed could
consequently affect trophic positioning (Tieszen et al. 1983).
Sampling designs need to incorporate a suitable level of replication to characterise the
trophic niche of higher consumers with variable feeding strategies. Yet post hoc
analyses determining suitable levels of replication are almost entirely absent from food
web studies utilising stable isotopes, despite the known high variability amongst
producers and consumers (Simenstad et al. 1993, Guest et al. 2010, Gillies et al. 2012).
95
Chapter IV
Further studies assessing stable isotope and diet variability across all feeding modes
amongst the benthos are necessary, in order to ensure the optimal allocation of resources
and to provide a more accurate picture of trophic niche, especially amongst the higherorder consumers.
Despite these limitations, trophic models derived by stable isotope analysis provide a
useful compromise, by summarising individual-level detail, in order to present
ecosystem-wide community models. Where possible, we increased the number of
replicates to ensure sufficient characterisation of the trophic niche of a species. Where
information was missing from this study, we used information derived from the
literature to ensure sufficient characterisation of each trophic category. Whilst we are
unable to determine quantitative estimates of carbon flow and individual contributions
for each producer, we were able to determine major and minor contributors of carbon
into the system.
Model components and description
Based on the stable isotope evidence from this study, we are able to propose a
generalised descriptive model for the shallow water benthic community of the Windmill
Islands (Figure 6). It is important to note that this model is incomplete, as some groups
are
under-represented
and
the
microbial
community
is
absent.
Notably,
microphytobenthic diatoms were not sampled, but are assumed to be an important
carbon source, with relatively enriched isotopic signatures similar to epiphytic diatoms
and sea ice algae (Corbisier et al. 2004). Despite these shortcomings, several important
details of carbon flow and trophic position can be inferred from the model.
Three major carbon pathways are available for the Windmill Islands benthic community
(Figure 6). The most depleted carbon chain (chain number one, figure 6) involves
pelagic POM, suspension feeders and their predators. Resuspended sediment POM
provides an additional carbon source to this chain. Selective feeding on different
particle sizes (sponges), and consumption of resuspended sediment POM, account for
the isotopic variation along the δ13C axis. Active incorporation of zooplankton,
particularly by holothurians, accounts for variation along the δ13N axis. We assume
ascidians, bryozoans and hydroids form part of this suspension-feeding guild based on
the isotopic values of Dunton (2001) and Jacob et al. (2006) who found the depleted
δ13C signatures in these groups closely matched phytoplankton. Specialised predators of
96
Chapter IV
suspension feeders include pycnogonids, nudibranchs and some asteroid species
(Dayton et al. 1974, Wägele 1989, McClintock 1994).
The second trophic pathway comprises a combination of macroalgae and epiphytic
algae,
which
is
incorporated
by grazers
and
their predators
(Figure 6).
Microphytobenthic diatoms are also assumed to be incorporated by some benthic
grazers. Sea ice diatoms are a secondary carbon sources for more mobile grazers (P.
walkeri). Few specialised grazers exist, with most species feeding on several sources.
Although the number of species that have a strictly herbivorous diet is small, largely
confined to several amphipods, isopods and gastropods (Arntz et al. 1994, Iken 1999,
Huang et al. 2007) herbivore biomass is high and provides an important food
component for top predators such as fish and penguins.
The third trophic pathway involves sediment POM, macroalgal detritus and benthic
diatoms, deposit feeders and their predators (Figure 6). The composition of sediment
POM is likely to fluctuate seasonally, based on benthic production and sea ice cover
over the summer months (Gibson et al. 1999, McMinn et al. 2004). Macroalgal-derived
POM is accessible throughout the year, although a time lag in its availability as a food
source is likely based on persistent chemical defences (Amsler et al. 2005). Deposit
feeders consist mainly of small crustaceans, polychaetes and several echinoderms.
Patchiness in the components of sediment POM, selective feeding and the inclusion of
microflora within the sediment, likely accounts for variation along the δ13C and δ15N
axis. Carbon is either retained within the sediment though small sediment-dwelling
predator/scavengers like H. videns and M. nordenskjoeldi, or transported to the
epibenthos via a scavenger/predator guild consisting of larger amphipod and isopod
crustaceans and fish.
97
Chapter IV
δ N
15
Figure 6. Trophic model of the Windmill Islands food web delineating the three major carbon pathways.
See text (subsection 4.4.2.) for carbon pathway descriptions. MA= Macroalgae, P= Pelagic POM, S=
Sediment POM, B/E D= Benthic/Epiphytic Diatoms, SI= Sea Ice POM. Dotted lines represent minor
contributions of carbon and solid lines represent major flows of carbon. X- and Y-axes represent the
boundaries of ecological niche space as defined by isotopic values.
98
Chapter IV
Conclusion
Benthic consumers in coastal waters of the Windmill Islands rely differentially on
several carbon sources which were readily distinguishable by δ13C isotopes. Analysis of
δ13C and δ15N revealed tight grouping of the suspension-feeding, grazer/herbivore and
deposit-feeding guilds, signifying that members within each guild rely on food sources
with
similar
carbon
signatures.
High-end
trophic
guilds
(predators
and
predator/scavengers) had a wide range of δ13C signatures and displayed little grouping,
reflecting considerable dietary variation. These results enabled us to determine that
carbon flow amongst the benthic fauna in the Windmill Islands is complex, with
multiple sources of carbon being utilised. However, these carbon flows can be
simplified into three main avenues for the movement of carbon from producers to
consumers. Discrete trophic levels were absent, with trophic organization more closely
representing a ‘trophic continuum’. Stable isotopes of carbon and nitrogen proved
useful in determining trophic guilds and identifying carbon sources, particularly
amongst first-order consumers, and were successful in determining trophic position
across all levels. We highlight the need for more specialist studies of diets, in
conjunction with stable isotope analysis and new techniques such as DNA analysis of
gut contents (e.g. Jarman et al. 2004, Dunshea 2009) to resolve trophic interactions
which cannot be delineated through stable isotope analysis alone.
99
Chapter IV
Acknowledgements
We thank members of the summer 2006/07 Casey dive and 2008/09 field teams for
assistance with field collections. We are very grateful for advice on stable isotope
preparations provided by Melissa Bautista and advice of food web studies provided by
Katrin Iken. This research was funded by a PhD research scholarship in part by
University of New England and Sothern Cross University, and supported financially
and logistically by the Australian Antarctic Division (AAS projects 2948 and 2201).
100
Chapter IV
References
Abrantes K, Sheaves M (2009) Food web structure in a near-pristine mangrove area of
the Australian Wet Tropics. Estuar Coast Shelf Sci 82:597-607
Ahn I (1997) Feeding ecology of the Antarctic lamellibranch Laturnula elliptica
(Laturnulidae) in Marian Cove and vicinity King George Island, during one austral
summer. In: Battaglia B, Valencia J, Walton DWH (eds) Antarctic communities:
species structure and survival. Cambridge University Press, Cambridge
Amsler CD, Iken K, McClintock J, Amsler MO, Peters KJ, Hubbard JM, Furrow FB,
Baker B (2005) Comprehensive evaluation of the palatability and chemical defences of
subtidal macroalgae from the Antarctic Peninsula. Mar Ecol Prog Ser 294:141-159
Anderson MJ (2001) A new method for non-parametric multivariate analysis of
variance. Austral Ecol 26:32-46
Arntz W, Brey T, Gallardo VA (1994) Antarctic zoobenthos. Oceanogr Mar Biol Annu
Rev 32:241-304
Barnes DKA, Clarke A (1995) Seasonality of feeding activity in Antarctic suspension
feeders Polar Biol 15:335-340
Blazewicz-Paszkowycz M, Ligowski R (2002) Diatoms as food source indicator for
some Antarctic Cumacea and Tanaidacea (Crustacea). Antarct Sci 14:11-15
Burns JM, Trumble SJ, Castellini MA, Testa JW (1998) The diet of Weddell seals in
McMurdo Sound, Antarctica as determined from scat collections and stable isotope
analysis. Polar Biol 19:272-282
Carabel S, Godinez-Dominguez E, Verisimo P, Fernandez L, Freire J (2006) An
assessment of sample processing methods for stable isotope analyses of marine food
webs. J Exp Mar Biol Ecol 336:254-261
Casaux R, Barrera-Oro E, Baroni A, Ramón A (2003) Ecology of inshore notothenioid
fish from the Danco Coast, Antarctic Peninsula. Polar Biol 26:157-165
Clarke A (1985) Energy flow in the Southern Ocean food web. In: Siegfried WR,
Condy PR, Laws RM (eds) Antarctic nutrient cycles and food webs. Springer-Verlag,
Berlin Heidelberg
101
Chapter IV
Corbisier T, Petti MV, Skowronski RP, Brito TS (2004) Trophic relationships in the
nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C stable-isotope
analysis. Polar Biol 27:75-82
Dauby P, Scailteur Y, De Broyer C (2001) Trophic diversity within the eastern Weddell
Sea amphipod community. Hydrobiologia 443:69-86
Dayton PK, Gordon AR, Paine RT, Dayton LB (1974) Biological accommodation in the
benthic community at McMurdo Sound, Antarctica. Ecol Monogr 44:105-128
Dayton PK, Mordida BJ, Bacon F (1994) Polar marine communities. Am Zool 34:90-99
Dayton PK, Oliver JS (1977) Antarctic soft-bottom benthos in oligotrophic and
eutrophic environments. Science 197:55-58
Dayton PK, Robilliard GA, Paine RT (1970) Benthic faunal zonation as a result of
Anchor Ice at McMurdo Sound, Antarctica. In: Holdgate M (ed) Antarctic Ecology,
Book 1. Academic Press, London
Dearborn JH, Hendler G, Edwards KC (1996) The diet of Ophiosparte gigas
(Echinodermata: Ophiuroidea) along the Antarctic Peninsula, with comments on its
taxonomic status. Polar Biol 16:309-320
Deniro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in
animals. Geochim Cosmochim Acta 42:495-506
Deniro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in
animals. Geochim Cosmochim Acta 45:341-351
Dhargalkar VK, Burton HR, Kirkwood JM (1988) Animal associations with the
dominant species of shallow water macrophytes along the coastline of the Vestfold
Hills, Antarctica. Hydrobiologia 165:141-150
Ducklow HW, Fraser W, Karl DM, Quetin LB, Ross RM, Smith RC, Stammerjohn SE,
Vernet M, Daniels RM (2006) Water-column processes in the West Antarctic Peninsula
and the Ross Sea: interannual variations and food web structure. Deep Sea Res Pt II
53:834-852
Daniels RM (2006) Water-column processes in the West Antarctic Peninsula and the
Ross Sea: Interannual variations and foodweb structure. Deep Sea Res Pt II 53:834-852
Dunshea G (2009) DNA-based diet analysis for any predator. PLoS ONE 4:4 e5252
102
Chapter IV
Dunton KH (2001) δ15N and δ13C measurements of Antarctic Peninsula fauna: trophic
relationships and assimilation of benthic seaweeds. Amer Zool 41:99-112
Fauchald K, Jumars PA (1979) The diet of worms: a study of polychaete feeding guilds.
Oceanogr Mar Biol Annu Rev 17:193-284
France RL (1995a) Carbon-13 enrichment in benthic compared to planktonic algae:
food web implications. Mar Ecol Prog Ser 124:307-312
France RL (1995b) Differentiation between littoral and pelagic food webs in lakes using
stable carbon isotopes. Limnol Oceanogr 40:1310-1313
France RL, Chandler M, Peters R (1998) Mapping trophic continua of benthic food
webs: body size-δ15N relationships. Mar Ecol Prog Ser 174:301-306
Fry B (1988) Food web structure on Georges Bank from stable C, N, and S isotopic
compositions. Limnol Oceanogr 33:1182-1190
Fry B (2006) Stable isotope ecology, Vol. Springer, New York
Fry B, Sherr EB (1989) 13C measurements as indicators of carbon flow in marine and
freshwater ecosystems. In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable Isotopes
in Ecological Research, Book 68. Springer-Verlag, New York
Gambi MC, Lorenti M, Russo GF, Scipione MB (1994) Benthic associations of the
shallow hard bottoms off Terra Nova Bay, Ross Sea: zonation, biomass and population
structure. Antarct Sci 6:449-462
Gibson JE, Trull T, Nichols PD, Summons RE, McMinn A (1999) Sedimentation of
13
C-rich organic matter from Antarctic sea-ice algae: a potential indicator of past sea-ice
extent. Geology 27:331-334
Gili JM, Coma R, Orejas C, López-González P, Zabala M (2001) Are Antarctic
suspension-feeding communities different from those elsewhere in the world? Polar
Biol 24:473-485
Gillies C, Stark J, Smith S (2012) Small-scale spatial variation of δ13C and δ15N
isotopes in Antarctic carbon sources and consumers. Polar Biol 1-15 In Press
Guest M, Hirst A, Nichols P, Frusher S (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
103
Chapter IV
Gutt J (1991) On the distribution and ecology of holothurians in the Weddell Sea
(Antarctica). Polar Biol 11:145-155
Gutt J, Sirenko BI, Smirnov IS, Arntz W (2004) How many macrozoobenthic species
might inhabit the Antarctic shelf? Antarct Sci 16:11-16
Huang Y, Amsler M, McClintock J, Amsler C, Baker B (2007) Patterns of
gammaridean amphipod abundance and species composition associated with dominant
subtidal macroalgae from the western Antarctic Peninsula. Polar Biol 30:1417-1430
Iken K (1999) Feeding ecology of the Antarctic herbivorous gastropod Laevilacunaria
antarctica Martens. J Exp Mar Biol Ecol 236:133-148
Iken K, Barrera-Oro ER, Quartino ML, Casaux RJ, Brey T (2004) Grazing by the
Antarctic fish Notothenia coriiceps: evidence for selective feeding on macroalgae.
Antarct Sci 9:386-391
Iken K, Bluhm BA, Gradinger R (2005) Food web structure in the high Arctic Canada
Basin: evidence from δ13C and δ15N analysis. Polar Biol 28:238-249
Iken K, Quartino M, Wiencke C (1999) Histological identification of macroalgae from
stomach contents of the Antarctic fish Notothenia coriiceps using semi-thin sections.
Mar Ecol 20:11-17
Isaacs JD (1973) Potential trophic biomasses and trace-substance concentrations in
unstructured marine food webs. Mar Biol 22:97-104
Jacob U, Brey T, Fetzer I, Kaehler S, Mintenbeck K, Dunton K, Beyer K, Struck U,
Pakhomov EA, Arntz WE (2006) Towards the trophic structure of the Bouvet Island
marine ecosystem. Polar Biol 29:106-113
Jacob U, Mintenbeck K, Brey T, Knust R, Beyer K (2005) Stable isotope food web
studies: a case for standardized sample treatment. Mar Ecol Prog Ser 287:251-253
Jarman SN, Deagle BE, Gales NJ (2004) Group-specific polymerase chain reaction for
DNA-based analysis of species diversity and identity in dietary samples. Mol Ecol
13:1313-1322
Johnston E, Connell S, Irving A, Pile A, Gillanders B (2007) Antarctic patterns of
shallow subtidal habitat and inhabitants in Wilke’s Land. Polar Biol 30:781-788
104
Chapter IV
Kaehler S, Pakhomov EA, McQuaid CD (2000) Trophic structure of the marine food
web at the Prince Edward Island (Southern Ocean) determined by 13C and 15N analysis.
Mar Ecol Prog Ser 208:13-20
Kennedy P, Kennedy H, Papadimitriou S (2005) The effect of acidification on the
determination of organic carbon, total nitrogen and their stable isotopic composition in
algae and marine sediment. Rapid Commun Mass Spectrom 19:1063-1068
Kiest KA (1993) A relationship of diet to prey abundance and the foraging behaviour of
Trematomus bernacchii. Polar Biol 13:291-296
La Mesa M, Dalú M, Vacchi M (2004) Trophic ecology of the emerald notothen
Trematomus bernacchii (Pisces, Nototheniidae) from Terra Nova Bay, Ross Sea,
Antarctica. Polar Biol 27:721-728
Lane A, Riddle MJ (2004) Toxicity of dispersed and undispersed diesel in sediments to
the Antarctic Ophiuroid Ophiura crassa. Australas J Ecotoxicol 10:115-121
Link J (2002) Does food web theory work for marine ecosystems? Mar Ecol Prog Ser
230:1-9
Logan JM, Lutcavage ME (2008) A comparison of carbon and nitrogen stable isotope
ratios of fish tissues following lipid extractions with non-polar and traditional
chloroform/methanol solvent systems. Rapid Commun Mass Spectrom 22:1081-1086
Massom RA, Stammerjohn SE (2010) Antarctic sea ice change and variability physical and ecological implications. Polar Sci 4:149-186
McArdle BH, Anderson MJ (2001) Fitting multivariate models to community data: a
comment on distance based redundancy analysis. Ecology 82:290-297
McClintock JB (1994) Trophic biology of Antarctic shallow-water echinoderms. Mar
Ecol Prog Ser 111:191-202
McClintock JB, Amsler CD, Baker BJ, Van Soest RWM (2005) Ecology of Antarctic
marine sponges: an overview. Integr Comp Biol 45:359-368
McCutchan JH, Lewis WM, Kendall C, McGrath CC (2003) Variation in trophic shift
for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102:378-390
105
Chapter IV
McMahon KW, Ambrose Jr WG, Johnson BJ, Sun M-Y, Lopez GR, Clough LM,
Carroll ML (2006) Benthic community response to ice algae and phytoplankton in Ny
Ålesund Svalbard. Mar Ecol Prog Ser 310:1-14
McMinn A, Runcie JW, Riddle MJ (2004) Effects of seasonal sea ice breakout on the
photosynthesis of benthic diatom mates at Casey, Antarctica. J Phycol 40:62-69
Minagawa M, Wada E (1984) Stepwise enrichment of
15
N along food chains: further
evidence and the relation between δ N and animal age. Geochim Cosmochim Acta
15
48:1135-1140
Mincks SL, Smith CR, Demaster DJ (2005) Persistence of labile organic matter and
microbial biomass in Antarctic shelf sediments: evidence of a sediment 'food bank'. Mar
Ecol Prog Ser 300:3-19
Mincks SL, Smith CR, Jeffreys RM, Sumida PYG (2008) Trophic structure on the West
Antarctic Peninsula shelf: Detritivory and benthic inertia revealed by δ13C and δ15N
analysis. Deep Sea Res Pt II 55:2502-2514
Murray MD, Luders DJ (1990) Faunistic studies at the Windmill Islands, Wilkes Land,
East Antarctica, 1959-80. ANARE Research Notes 73:45pp
Norkko A, Thrush SF, Cummings V, Gibbs HL, Andrew N, Norkko J, Schwarz A-M
(2007) Trophic structure of coastal Antarctic food webs associated with changes in sea
ice and food supply. Ecology 88:2810-2820
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 softsediment communities of McMurdo Sound, Antarctica. Polar Biol 27:482-494
Nyssen F, Brey T, Lepoint G, Bouquegneau J-M, De Broyer C, Dauby P (2002) A
stable isotope approach to the eastern Weddell Sea trophic web: focus on benthic
amphipods. Polar Biol 25:280-287
Oliver JS, Slattery PN (1985) Effects of crustacean predators on species composition
and population structure of soft-bodied infauna from McMurdo Sound, Antarctica.
Ophelia 24:155-175
Orejas C, Gili J, López-González P, Arntz W (2001) Feeding strategies and diet
composition of four Antarctic cnidarian species Polar Biol 24:620-627
106
Chapter IV
Pearse JS, Giese AC (1966) Food, reproduction and organic constitution of the
Antarctic echinoid Sterechinus neumayeri (Meissner). Biol Bull 130:387-401
Peterson BJ (1999) Stable isotopes as tracers of organic matter input and transfer in
benthic food webs: a review. Acta Oecol 20:479-487
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst
18:293-320
Phillips D (2001) Mixing models in analyses of diet using multiple stable isotopes: a
critique. Oecologia 127:166-170
Polis GA, Strong DR (1996) Food Web Complexity and Community Dynamics. The
American Naturalist 147:813-846
Post D, Layman C, Arrington D, Takimoto G, Quattrochi J, Montaña C (2007) Getting
to the fat of the matter: models, methods and assumptions for dealing with lipids in
stable isotope analyses. Oecologia 152:179-189
Post DM (2002) Using stable isotopes to estimate trophic position: models, methods,
and assumptions. Ecology 83:703-718
Purinton BL, Demaster DJ, Thomas CJ, Smith CR (2008)
14
C as a tracer of labile
organic matter in Antarctic benthic food webs. Deep Sea Res II 55:2438-2450
Quartino M, Boraso De Zaixso A (2008) Summer macroalgal biomass in Potter Cove,
South Shetland Islands, Antarctica: its production and flux to the ecosystem. Polar Biol
31:281-294
Rakusa-Suszczewski S (1972) The biology of Paramoera walkerii Stebbing
(Amphipoda) and the Antarctic sub-fast ice community. Polskie Archiwum
Hydrobiologii 19:11-36
Rau GH, Sweeney RE, Kaplan IR (1982) Plankton 13C:12C ratio changes with latitude:
differences between northern and southern oceans. Deep Sea Res Pt I 29:1035-1039
Rau GH, Teyssie JL, Rassoulzadegan F, Fowler SW (1990)
13
C /12C and
15
N/14N
variations among size fractionated marine particles: implications for their origin and
trophic relationships. Mar Ecol Prog Ser 59:33-38
Rounick JS, Winterbourn MJ (1986) Stable carbon isotopes and carbon flow in
ecosystems. Biosci 36:171-177
107
Chapter IV
Sackett WM, Eckelmann WR, Bender ML, Be AWH (1965) Temperature dependence
of carbon isotope composition in marine plankton and sediments. Science 148:235-237
Simenstad CA, Duggins DO, Quay PD (1993) High turnover of inorganic carbon in
kelp habitats as a cause of δ13C variability in marine food webs. Mar Biol 116:147-160
Smale DA, Barnes DKA, Fraser KPP, Mann PJ, Brown MP (2007) Scavenging in
Antarctica: intense variation between sites and seasons in shallow benthic necrophagy. J
Exp Mar Biol Ecol 349:405-417
Smith CR, Mincks S, Demaster DJ (2006) A synthesis of bentho-pelagic coupling on
the Antarctic shelf: Food banks, ecosystem inertia and global climate change. Deep Sea
Res Pt II 53:875-894
Snape I, Riddle MJ, Stark JS, King CK, Duquesne S, Gore DB (2001) Management and
remediation of contaminated sites at Casey Station, Antarctica. Polar Rec 37:199-214
Søreide JE, Hop H, Carroll ML, Falk-Petersen S, Hegseth EN (2006) Seasonal food
web structures and sympagic-pelagic coupling in the European Arctic revealed by stable
isotopes and a two-source food web model. Prog Oceanogr 71:59-87
Stark JS (2000) The distribution and abundance of soft-sediment macrobenthos around
Casey Station, East Antarctica. Polar Biol 23:840-850
Tamelander T, Renaud PE, Hop H, Carroll ML, Ambrose Jr WG, Hobson KA (2006)
Trophic relationships and pelagic–benthic coupling during summer in the Barents Sea
Marginal Ice Zone, revealed by stable carbon and nitrogen isotope measurements. Mar
Ecol Prog Ser 310:33-46
Tatián M, Sahade R, Mercuri G, Fuentes V, Antacli J, Stellfeldt A, Esnal G (2008)
Feeding ecology of benthic filter-feeders at Potter Cove an Antarctic coastal ecosystem.
Polar Bio 31:509-517
Thomas D, Jiang J (1986) Epiphytic diatoms of the inshore marine area near Davis
Station. Hydrobiologia 140:193-198
Thompson DR, Riddle MJ (2005) Bioturbation behaviour of the spatangoid urchin
Abatus ingens in Antarctic marine sediments. Mar Ecol Prog Ser 290:135-143
Thurber A (20070 Diets of Antarctic sponges: links between the pelagic microbial loop
and benthic metazoan food web. Mar Ecol Prog Ser 351:77-89
108
Chapter IV
Tieszen LL, Boutton TW, Tesdahl KG, Slade NA (1983) Fractionation and turnover of
stable carbon isotopes in animal tissues: Implications for δ13C analysis of diet.
Oecologia 57:32-37
Vander Zanden MJ, Rasmussen JB (1999) Primary consumer δ13C and δ15N and the
trophic position of aquatic consumers. Ecology 80:1395-1404
Vander Zanden MJ, Rasmussen JB (2001) Variation in δ15N and δ13C trophic
fractionation: Implications for aquatic food web studies. Limnol Oceanogr 46:20612066
Vanderklift M, Ponsard S (2003) Sources of variation in consumer-diet δ15N
enrichment: a meta-analysis. Oecologia 136:169-182
Wada E, Terazaki M, Kabaya Y, Nemoto T (1987)
15
N and
13
C abundances in the
Antarctic Ocean with emphasis on the biogeochemical structure of the food web. Deep
Sea Res Pt I 34:829-841
Wägele H (1989) Diet of some Antarctic nudibranchs (Gastropoda, Opisthobranchia,
Nudibranchia). Mar Biol 100:439-441
Wiencke C, Clayton MN, Gómez I, Iken K, Lüder UH, Amsler CD, Karsten U, Hanelt
D, Bischof K, Dunton K (2007) Life strategy ecophysiology and ecology of seaweeds in
polar waters. Rev Environ Sci Biotechnol 6:95-126
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 23:107-1
109
Chapter V
110
Chapter V
Chapter V
Food web structure of the coastal benthic community,
Vestfold Hills: similarity within East Antarctica
Christopher L. Gillies, Jonathan S. Stark, Glenn J. Johnstone and Stephen DA. Smith
A revised version has been submitted to Marine Ecology Progress Series
111
Chapter V
Abstract
Stable isotopes of carbon and nitrogen were used to identify carbon sources and trophic
position in the high-latitude, shallow water, benthic community of the Vestfold Hills,
East Antarctica. Carbon sources were generally well separated by δ13C and ranged from
-30.13‰ for the red alga Gymnogongrus to -8.79‰ for sea ice POM. Lower-order
consumers could be grouped according to their feeding guild and main dietary sources
as determined by δ13C and δ15N. Higher-order consumers occupied the full range of
δ13C ratios and had similar δ15N signatures, although predators were weakly, although
significantly, enriched in δ15N compared to scavenger/predators and omnivores. We
then compared the Vestfold Hills food web to the food web model previously described
for the Windmill Islands. We found similar δ13C ratios for several co-occurring carbon
sources, whilst the δ15N ratios in consumers from the Vestfold Hills were consistently
enriched compared to those from the Windmill Islands by 1-2‰. The relative position
of feeding guilds on the δ13C and δ15N plane were similar for both food webs,
demonstrating that stable isotopes of carbon and nitrogen are successful in repeating
consistent patterns of carbon flow and trophic position for similar communities located
1000s of km apart.
Key words: δ13C, Casey Station, Davis Station, δ15N, stable isotopes, trophic ecology,
Windmill Islands
112
Chapter V
Introduction
Understanding carbon flow and trophic linkages in the context of a food web is a
fundamental requisite in determining ecosystem-wide changes to community structure
and function (Post 2002a, Polis 1994). The need for such understanding is particularly
imperative in polar environments where modifications in climate are expected to bring
about large-scale changes to the environment. For example, changes to air and water
temperatures in the Antarctic, induced by climate change, are likely to bring about
altered sea ice cover and consequently affect light regimes and primary productivity
(Stammerjohn et al. 2008). Bottom-up changes in carbon resources may lead to direct
shifts in the distribution and abundance of benthic fauna, significantly altering the
structure and function of benthic communities (Massom and Stammerjohn 2010).
Understanding the links between carbon sources and consumers is also central to
detecting bottom-up changes to contamination pathways (Hobson et al. 2002,
Macdonald et al. 2005) and determining the top-down effects of the removal of large
predators (Barbraud and Weimerskirch 2001).
Food webs in Antarctic nearshore benthic communities appear to be characterised by a
high level of consumer connectivity, indicated by a high degree of omnivory, in
response to strong seasonality in the timing and quantity of primary production (Arntz
et al. 1994, Dayton et al. 1994, Barnes and Clarke 1995). Feeding plasticity and lack of
specialised feeders in response to variable carbon sources point towards considerable
food web complexity (Jacob et al. 2006, Norkko et al. 2007). A lack of discernible
trophic levels (Kaehler et al. 2000, Corbisier et al. 2004, Gillies et al. 2012b), suggests
shallow-water benthic food webs are more representative of a trophic ‘continua’ (France
et al. 1998) rather than traditional step-wise trophic models (Hairston et al. 1960,
Hairston and Hairston 1993). Thus Antarctic nearshore food webs may represent
relatively stable systems, summarised by strong connectivity and weak interactions
between food web members (McCann 2000). Yet currently, our ability to identify
generalised patterns in food web structure for the Antarctic nearshore region is
considerably hampered by: (1) the paucity of descriptions of coastal food webs or
carbon flow from high-latitude areas and regions outside of the Antarctic Peninsula or
the Ross Sea and; (2) the lack of comparative studies.
113
Chapter V
No study to date has identified consistent patterns in benthic food web dynamics across
similar communities from different regions in Antarctica. This is despite the generation
of several models describing carbon flow and trophic associations for nearshore benthic
communities (Dunton 2001, Corbisier et al. 2004, Jacob et al. 2006) and a growing
body of literature indicating that Antarctic nearshore food webs share similar patterns of
resource use and trophic width (Dayton et al. 1974, Dayton and Oliver 1977, Kaehler et
al. 2000, Dunton 2001, Corbisier et al. 2004, Jacob et al. 2006). The current lack of
comparative studies limits our ability to model the effects of large-scale impacts beyond
the regional scale and, therefore, identify consistent patterns brought about by largescale processes. Additionally, the identification of spatial consistencies amongst food
webs from different regions will allow greater insight into the application of food web
theory to Antarctic systems and consequently allow studies from the Antarctic to
contribute to food web theory.
The effectiveness of analysing stable isotopes of carbon and nitrogen to describe carbon
flow, and time-integrated trophic positions under polar conditions, has been
successfully demonstrated in several studies (e.g. Hobson et al. 1995, Dunton 2001,
Iken et al. 2005, Jacob et al. 2006). The basis of using isotopic measurements to study
trophic structure lies in documenting a regular and consistent pattern of isotopic
enrichment with increasing trophic level (Fry 1988). Nitrogen isotope ratios (14N:15N,
expressed as δ15N) in consumers become enriched by 3-4 ‰ with each trophic level,
enabling elucidation of trophic position (Deniro and Epstein 1981, Minagawa and Wada
1984). Carbon signatures (12C:13C, expressed as δ13C) remain relatively stable amongst
trophic levels, thus enabling various carbon sources at the base of the food web to be
linked with higher-order consumers (DeNiro and Epstein 1978, Rounick and
Winterbourn 1986, Peterson and Fry 1987). Identification and partitioning of individual
carbon sources is possible when end members have sufficiently separate isotopic
signatures (Fry and Sherr 1989, Peterson 1999).
We used stable isotopes of carbon and nitrogen to identify carbon flows and trophic
structure within the shallow-water benthic ecosystem of the Vestfold Hills. Our primary
goal was to build a food web model derived from stable isotopes of carbon and nitrogen
and to: (1) identify the main carbon sources utilised by the benthic fauna; and (2)
provide estimates of trophic position. We then tested the generality of our model with
the isotopic model developed for the Windmill Island food web (Gillies et al. 2012b) by
114
Chapter V
comparing the isotopic signatures of co-occurring carbon sources and consumers and
the isotopic position of each trophic guild within the food web.
Figure 1. Location of study regions (A) Vestfold Hills and (B) Windmill Islands and sampling sites (
115
).
Chapter V
Methods
Study location
The nearshore area surrounding the Vestfold Hills region in Prydz Bay, East Antarctica
(Figure 1) supports a rich benthic community (Everitt et al. 1980, Tucker and Burton
1987, Dhargalkar 1990), comparable with that of similar habitats from other parts of
Antarctica (Tucker and Burton 1988). The rocky shores provide an important breeding
and haul-out area for seals and birds during the summer months, whilst the coastal area
is covered in thick fast-ice over the winter period. Strong seasonality in primary
production occurs in both pelagic and benthic habitats (Tucker and Burton 1988), whilst
organic matter derived from sea ice algae can comprise up to 20% of the total particle
flux reaching the benthos over the summer months (Gibson et al. 1999).
Sampling design
We sampled several benthic macroalgae, in addition to pelagic, sediment and sea ice
particulate organic matter (POM), and consumers from several different locations
occupying the shallow, nearshore zone (0-30 m) around the Vestfold Hills, in the
summer season of 2009/10. We aimed to collect many species across a range of feeding
guilds in order to describe carbon flows and trophic pathways in as much detail as
possible amongst the benthic community. As recommended by Gillies et al. (2012a),
consumers were collected as close to their potential food sources.
Carbon sources
Sediment POM samples were collected by divers at depths of 6-30 m by coring (5 cm
diameter by 10 cm long) and utilising only the top 1.5 cm section of the core. Larger
infauna were removed prior to drying and an homogenised sediment POM sub-sample
was analysed. Macroalgal samples were collected by divers and kept dark until frozen
for analysis. Pelagic POM samples were collected by monthly (January-March)
horizontal tows conducted at the surface, adjacent to the coast (53 µm mesh, 100 m
from shore). Samples were dominated by the diatom Trigonium antarcticum and
dinoflagellates. Sea ice POM samples were collected from fast ice adjacent to the coast
by SIPRE corer and comprised a mixture of diatoms and bacteria. Both pelagic POM
and sea ice POM samples were kept dark in the field at approx. 4°C until they were
116
Chapter V
brought back to the laboratory. Samples were then scanned under a dissecting
microscope to remove large zooplankton and spun in a centrifuge at 3400 rpm for 10
min: the supernatant was discarded and the remaining sample rinsed with Milli-Q water,
re-spun, dried and frozen at -20°C for subsequent analysis.
Consumers
All consumers with the exception of fish, tanaids, amphipods and nephtyid polychaetes,
were collected by divers at depths between 2-30 m and by snorkelling at depths down to
2 m. Fish were collected with a long line or in baited traps between 8-30 m. Tanaids,
amphipods and nephtyid polychaetes were either sieved from the sediment cores,
collected from the fish traps (amphipods) or collected using a dip net in waters <2 m.
All samples were frozen at -20°C until analysis. Reference specimens of ascidians,
holothurians and sponges are held at the Australian Antarctic Division, Hobart.
Stable isotope analysis
Several tissue types were selected for stable isotope analysis amongst the benthic fauna
(Table 1). Where possible, muscle tissue was prioritised, following by epidermal tissue,
body sections and whole organisms (gut contents not removed). All samples were
thoroughly rinsed in Milli-Q water prior to being dried in an oven at 60°C for 48 hours
and ground to a fine powder. Carbonates were removed from those samples where it
was not possible to obtain carbonate-free tissue (asteroids, holothurians, ophuroids,
echinoids, crustaceans, and sediment POM) using 1M HCL and the ‘drop-by-drop’
method recommended by Jacob et al. (2005) and Carabel et al. (2006).
Lipids can be depleted in δ13C by up to 9‰ relative to muscle tissue (DeNiro and
Epstein 1977, McConnaughey and McRoy 1979). Consequently, to standardise lipid
content amongst different tissue types, a post-hoc mathematical normalisation formula
was applied to δ13C values in samples where lipid content was high (Post et al. 2007).
We used the C:N ratio of samples as a proxy for lipid content, and where C:N exceeded
3.5 (i.e. high lipid content), applied the formula: ∆13C = –3.32 + 0.99 x C:N (Post et al.
2007). The mean difference between δ13Clipid and δ13Cnorm were: 0.95‰ ± 0.05 (± 1
standard error, n=229).
Prior to the main stable isotope analysis, and due to the difficulty in obtaining low or
carbonate-free tissue samples, we conducted a pilot study to determine δ13C similarity
117
Chapter V
between different tissue types in asteroids. We noted an increasing enrichment of δ13C
amongst different tissue types after carbonate extraction and lipid normalisation
(epidermis>tube feet>intestinal tissue, C. Gillies unpublished data). We therefore report
δ13C values of asteroid tissue (and potentially echinoids and ophuroids) with the caveat
that epidermal δ13C values may be enriched by 2.14‰ to 3.20‰ compared to other
tissues, even despite carbonate removal and lipid normalisation. For δ15N, intestinal
tissue was approx. 2‰ depleted compared to epidermal and tube feet tissue, which
shared similar δ15N values. We chose to use epidermal tissue in this study for
consistency with other Antarctic studies that have used epidermal/body wall tissue in
echinoderms (Jacob et al. 2006, Norkko et al. 2007, Mincks et al. 2008, Gillies et al.
2012b) but highlight the need for further investigation into isotope variability between
different tissue types for Antarctic echinoderms.
Carbon (δ13C) and Nitrogen (δ15N) signatures were analysed by a continuous-flow
stable isotope ratio mass spectrometer (Fisons Isochrom) coupled to an element analyser
at the Research School of Biology, Australian National University, Canberra. The
working standards used for δ13C were C3 beet sucrose and C4 cane, calibrated against
the global standard VPDB, whilst several synthetic standards were used for δ15N,
calibrated against the global standard of atmospheric nitrogen. Standard deviations of
replicate samples (n=149) were 0.09‰ for δ13C and 0.14‰ for δ15N.
Results are expressed in the standard delta notation:
δX= ([Rsample/Rstandard-1)*1000
where X = Carbon or Nitrogen and R = the ratio of the heavy isotope over the light
isotope.
Data analysis
Animals were separated by taxon and feeding guilds and plotted in δ13C versus δ15N biplots to visualise differences in diet and trophic position. We used the δ15N values of the
suspension-feeder Laternula elliptica 4 as a baseline estimate of trophic position (Gillies
et al. 2012a), as long-living, sedentary, first-order species are best suited to act as
4
We used Laternula elliptica here instead of Adamussium colbecki because it was far more common in
the Vestfold Hills that A. colbecki, which we deemed to be more appropriate for baseline analysis. Both
species had almost identical isotopic values, resulting in no difference to trophic analysis when using
either species.
118
Chapter V
trophic baselines (Cabana and Rasmussen 1996, Post 2002b). We assumed a δ15N
trophic fractionation of 3‰, based on previous work (Vander Zanden and Rasmussen
2001, Post 2002b, McCutchan et al. 2003, Vanderklift and Ponsard 2003).We calculated
a δ13C fractionation of 0.1‰ per tropic level based on the mean difference between
suspension feeders and pelagic POM from this study.
The δ13C and δ15N signatures of co-occurring carbon sources and consumer species that
are common to both the Vestfold Hills and Windmill Islands were compared
statistically using a t-test (Quinn and Keough 2002). Furthermore, we compared our
isotopic bi-plot to the trophic model produced in Gillies et al. (2012b) and identified
similarities in trophic position of common food web members and the position of
feeding guilds.
119
Chapter V
Table 1. δ13C and δ15N values (mean ± SE) of carbon sources and consumer from the Vestfold Hills food
web. PP= primary producer, SF= suspension feeder, DF= deposit feeder, G=grazer, O=omnivore, S/P=
scavenger/predator, P= predator. *See Appendix 1 for authors to feeding guilds.
n
δ13C (‰)
δ15N (‰)
*
Feeding
guild
Tissue
analysed
Chaetomorpha sp.
6
-16.64 ± 0.15
5.88 ± 0.07
PP
whole frond
Monostroma sp.
6
-17.70 ± 0.79
6.74 ± 0.06
PP
whole frond
Phaeophyceae
Himantothallus
grandifolius
6
-17.47 ± 0.41
5.36 ± 0.20
PP
1cm2 sections
Rhodophyta
Iridaea cordata
18
-17.10 ± 0.35
6.80 ± 0.14
PP
1cm2 sections
Palmaria decipiens
6
-16.33 ± 0.30
6.32 ± 0.06
PP
whole frond
Gymnogongrus
1
-30.13
4.58
PP
whole frond
Pelagic POM
Pelagic POM
14
-23.08 ± 0.84
5.45 ± 0.14
surface tow
Sea Ice POM
Sea Ice POM
47
-8.79 ± 0.63
2.95 ± 0.14
Ice core
Sediment POM
Sediment POM
19
-18.68 ± 0.27
5.75 ± 0.21
sediment core
Flabelligera mundata
6
-15.30 ± 0.58
6.54 ± 0.29
DF
body section
Nephtyidae
3
-13.52 ± 0.76
11.92 ± 0.14
P
body section
Perkinsiana sp.
8
-19.27 ± 0.28
6.35 ± 0.19
SF
body section
Polynoidae
2
-14.42 ± 1.24
11.40 ± 0.52
P
body section
Terrebellidae
2
-20.61 ± 0.03
7.79 ± 0.29
DF
body section
Heterophoxus videns
2
-14.54 ± 0.08
13.76 ± 1.22
P
whole,
combined
individuals
Orchomenella franklini
1
-14.70
9.42
DF
whole,
combined
individuals
Paramoera walkeri
4
-14.19 ± 0.24
7.26 ± 0.28
G
whole,
combined
individuals
Tryphosella sp.
3
-17.79 ± 0.09
10.87 ± 0.14
S/P
whole
individual
Waldeckia obesa
3
-18.79 ± 0.21
10.23 ± 0.12
S/P
whole,
combined
individuals
Decapoda
Chorismus antarcticus
2
-15.19 ± 1.00
10.80 ± 0.43
O
muscle tissue
Isopoda
Cymodocella tubicauda
2
-11.94 ± 0.61
6.14 ± 0.38
G
whole,
combined
individuals
Species
Chlorophyla
Annelida
Polychaeta
Arthropoda
Amphipoda
120
Chapter V
Pycnogonida
Tanaidacea
Glyptonotus antarcticus
9
-13.93 ± 0.59
12.44 ± 0.25
P
muscle tissue
Arcturidae
2
-16.50 ± 0.13
6.10 ± 0.02
SF
whole
individual
Magnammothea gigantea
2
-17.56 ± 0.08
11.93 ± 1.51
P
body sectionleg
Nymphon australe
8
-16.97 ± 0.45
10.39 ± 0.26
P
whole
individual
Tanaids
1
-9.09
8.27
DF
whole,
combined
individuals
Ascidian sp. 1
1
-17.92
2.47
SF
body section
Ascidian sp. 2
3
-22.70 ± 0.02
6.10 ± 1.02
SF
body section
Ascidian sp. 3
3
-17.50 ± 0.24
6.39 ± 0.12
SF
body section
Ascidian sp. 4
1
-19.87
7.45
SF
body section
Ascidian sp. 5
1
-20.70
6.70
SF
body section
Ascidian sp. 6
1
-20.29
6.47
SF
body section
Ascidian sp. 7
1
-19.93
7.11
SF
body section
Ascidian sp. 8
1
-19.92
8.34
SF
body section
Ascidian sp. 9
1
-18.77
6.71
SF
body section
Ascidian sp10
1
-17.54
3.36
SF
body section
Ascidian sp11
2
-21.92 ± 1.90
6.88 ± 1.06
SF
body section
Ascidian sp. 12
16
-19.41 ± 0.41
7.01 ± 0.18
SF
body section
Ascidian sp. 13
1
-15.60
6.50
SF
body section
Ascidian sp. 14
3
-22.83 ± 0.22
4.74 ± 0.24
SF
body section
Pyrosoma sp.
1
-19.60
6.77
SF
body section
Chinodraco hamatus
13
-22.62 ± 0.14
12.30 ± 0.15
P
muscle tissue
Pagothina borchgrevinki
1
-21.59
10.89
P
muscle tissue
Trematomus bernacchii
131
-16.07 ± 0.08
13.98 ± 0.06
P
muscle tissue
Trematomus newnesi
1
-23.86
9.67
P
muscle tissue
Artemidactis victrix
3
-15.97 ± 0.27
12.02 ± 0.38
P
Epidermis
Isotealia antarctica
1
-17.07
12.84
P
Epidermis
Stomphia selaginella
5
-17.16 ± 0.43
11.56 ± 0.39
P
Epidermis
Urticinopsis antarctica
11
-15.64 ± 0.42
11.53 ± 0.29
P
Epidermis
Chordata
Ascidiacea
Perciformes
Cnidaria
Anthozoa
121
Chapter V
Hydrozoa
Pennatulacea
2
-20.76 ± 0.30
8.79 ± 0.41
SF
body section
Hydroid sp. 1
3
-21.98 ± 0.03
6.16 ± 0.16
SF
colony section
Hydroid sp. 2
4
-21.95 ± 0.31
6.27 ± 0.09
SF
colony section
Aconodaster cf. hodgsoni
3
-11.22 ± 0.52
11.72 ± 0.54
P
Epidermis
Aconodaster cf. conspicus
3
-15.02 ± 0.81
10.36 ± 0.41
P
Epidermis
Diplasterias brucei
11
-13.32 ± 0.45
10.14 ± 0.20
S/P
Epidermis
Odontaster validus
9
-11.32 ± 0.87
11.93 ± 0.18
O
Epidermis
Perkinaster aurorae
3
-15.07 ± 0.87
11.68 ± 0.27
P
Epidermis
7
-14.47 ± 0.49
10.91 ± 0.31
S/P
Epidermis
4
-11.93 ± 0.54
10.41 ± 0.15
S/P
Epidermis
Echinodermata
Asteroidea
Perkinaster
antarcticus
cf.
fuscus
Perkinaster cf.
antarcticus (juv)
fuscus
Echinoidea
Sterechinus neumayeri
12
-12.36 ± 0.41
10.40 ± 0.11
O
peristomal
membrane
Holothuroidea
Cucumaria sp. 1
3
-19.88 ± 0.34
7.99 ± 0.55
SF
body section
Cucumaria sp. 2
1
-21.71
8.08
SF
body section
Cucumaria sp. 3
2
-20.36 ± 0.22
9.17 ± 0.36
SF
body section
Cucumaria sp. 4
9
-20.92 ± 0.15
8.14 ± 0.18
SF
body section
Cucumaria sp. 5
5
-17.39 ± 1.17
8.22 ± 0.41
SF
body section
Cucumaria sp. 6
1
-20.16
8.41
SF
body section
Staurocucumis sp.
5
-19.81 ± 0.29
8.96 ± 0.49
SF
body section
Ophiosparte gigas
3
-14.90 ± 0.76
9.89 ± 0.43
P
Epidermis
Ophiura crassa
3
-15.81 ± 0.63
9.98 ± 0.35
DF
Epidermis
Adamussium colbecki
2
-19.69 ± 0.16
6.80 ± 0.19
SF
muscle tissue
Laternula elliptica
8
-19.71 ± 0.13
6.83 ± 0.09
SF
muscle tissue
Amauropsis rossiana
4
-19.10 ± 0.09
8.10 ± 0.07
P
muscle tissue
Marseniopsis mollis
8
-20.17 ± 0.20
9.68 ± 0.18
P
muscle tissue
Neobuccinium eatoni
18
-15.16 ± 0.29
12.07 ± 0.13
P
muscle tissue
Trophon longstaffi
5
-17.87 ± 0.26
9.68 ± 0.16
P
muscle tissue
Doris kerguelensis
9
-13.74 ± 0.43
10.87 ± 0.22
P
muscle tissue
Tritonia challengeriana
1
-20.14
11.32
P
muscle tissue
Cephalaspidea
4
-10.46 ± 0.13
9.36 ± 0.20
DF
muscle tissue
Ophuroidea
Mollusca
Bivalvia
Gastropoda
122
Chapter V
Nemertea
Parborlasia corrugatus
5
-16.19 ± 0.25
11.16 ± 0.07
S/P
body section
homaxinella sp.
5
-20.30 ± 0.26
6.66 ± 0.21
SF
body section
Isodictya sp.
4
-19.12 ± 0.43
7.14 ± 0.22
SF
body section
Kirkpatrickia sp.
1
-19.09
6.88
SF
body section
Sphaerotylus sp.
2
-19.29 ± 0.57
5.92 ± 0.99
SF
body section
Porifera sp. 1
1
-20.81
6.85
SF
body section
Porifera sp. 2
3
-18.60 ± 0.23
7.95 ± 0.69
SF
body section
Porifera sp. 3
1
-20.00
6.13
SF
body section
Porifera sp. 4
1
-20.42
6.72
SF
body section
Porifera sp. 5
1
-18.45
5.98
SF
body section
Porifera sp. 6
1
-20.07
6.71
SF
body section
Porifera sp. 7
2
-19.93 ± 0.42
7.90 ± 0.60
SF
body section
Porifera sp. 8
3
-20.74 ± 0.28
7.63 ± 0.29
SF
body section
Porifera
Demospongia
123
Chapter V
Results
Vestfold Hills food web
δ13C and δ15N in carbon sources
All carbon sources showed considerably different isotopic signatures, with pelagic
POM, sediment POM and sea ice POM being well separated in δ13C and δ15N from
benthic macroalgae (Table 1, Figures 2, 6a). The red alga Gymnogongrus sp. had the
least enriched δ13C signature at -30.13‰, followed by pelagic POM (-19.86‰) and
sediment POM (-18.68‰) whilst sea ice POM had the most enriched signature (8.79‰). The remaining five macroalgal species, consisting of two red, two green and
one brown algae, shared similar δ13C signatures ranging from -17.70‰ for the green
alga Monostroma sp. to -16.33‰ for the red alga Palmaria decipiens. With the
exception of sea ice POM (2.95‰) and Gymnogongrus sp. (4.58‰), all carbon sources
had considerably enriched δ15N signatures for producers relative to atmospheric
nitrogen, ranging from 5.36‰ for the brown algae Himantothallus grandifolius to
6.80‰ for the red algae Iridaea cordata (Table 1, Figures 3, 6a).
δ13C in consumers
Eighty-two consumer species were sampled, covering a wide range of taxonomic
groups and feeding guilds (Table 1). The δ13C ratios amongst all fauna spanned a range
of 13.74‰ and all consumers were within the δ13C range of sampled carbon sources.
The δ13C signatures of first-order consumers were aligned with their respective guilds,
although there was considerable variation within each guild (Figure 2). Among
suspension feeders, δ13C values ranged from -22.83‰ to -15.60‰, and were more
closely aligned to the pelagic POM carbon signature than to macroalgae (Figures 2, 6a).
However, several species displayed enriched carbon signatures beyond the δ13C ratios
of pelagic POM, most notably, several species of ascidians and poriferans (Figure 2).
Deposit feeders spanned a considerable δ13C range, from -20.61‰ for terrebellid
polychaetes, to -10.46‰ for Cephalaspidea and -9.09‰ for tanaids. The remaining
deposit feeders (Flabelligera mundata, Ophiura crassa, Orchomenella franklini), had
similar δ13C signatures (-14.70‰ to -15.81‰), which were enriched by approx. 3‰
over sediment POM. The two grazer species, Paramoera walkeri and Cymodocella
tubicauda, had δ13C signatures enriched by 2-4‰ over macroalgae.
124
Chapter V
The three omnivore species were relatively enriched in δ13C, closely matching benthic
carbon sources (Figure 2); however, their δ13C signatures were also similar to benthic
grazers. Scavenger/predators spanned a considerable δ13C range from -18.79‰ for the
amphipod Waldeckia obesa, to -11.93‰ for Perkinaster cf. focus antarcticus (juvenile).
On the whole, δ13C values across scavenger/predators, were more representative of
relatively enriched benthic carbon sources.
Predators spanned the full δ13C spectrum, indicating carbon sources derived from both
benthic and pelagic sources and a wide range of diets (Table 1, Figure 2). Notably, the
three fish species, Trematomus newensi, Chinodraco hamatus and Pagothina
borchgrevinki, were the most depleted of all predators ranging from -23.86‰ for T.
newensi to -21.59‰ for P. borchgrevinki. In contrast, T. bernacchii had a more
enriched signature (-16.07‰). Several predators likely to feed exclusively on members
of the suspension-feeding guild (e.g. Tritonia challengeriana, Marseniopsis mollis,
Amauropsis rossiana), also had considerably depleted δ13C values. Several asteroid
species and nephtyid polychaetes were comparatively enriched in δ13C, suggesting a
strong dependence on benthic carbon sources.
δ15N in consumers
Nitrogen values amongst all consumers spanned a δ15N range of 11.51‰, from 2.47‰
for Ascidian sp. 1 to 13.98‰ for the fish Trematomus bernacchii. In addition,
considerable variation occurred within several feeding guilds, with δ15N values among
members of the suspension-feeding, deposit-feeding, scavenger/predators and predator
guilds, spanning more than 3‰, equivalent to one trophic level (Figure 3).
Based on the assumed 3‰ enrichment per trophic level, the suspension-feeding guild
spanned 6.17‰ or two trophic levels (Figure 3). A considerable proportion of
suspension feeders had δ15N signatures greater than the designated trophic baseline (L.
elliptica, 6.83‰), notably several species of Cucumaria and Porifera, Pennatulacea and
Staurocucumis sp., with Cucumaria sp. 3 having the highest δ15N value amongst
suspension feeders (9.17‰). Grazers and deposit feeders were restricted to the second
trophic level (first-order consumers) although deposit feeders spanned a considerable
range from 6.54‰ for F. mundata to 9.98‰ for Ophiura crassa.
Predators, predator/scavengers and omnivores occupied the third trophic level as
second-order consumers and showed little trophic separation (Figures 3, 6a). However,
125
Chapter V
statistical analysis revealed the predator guild (mean 12.74‰) was slightly, although
significantly, higher compared to predator/scavengers and omnivores (one-way
ANOVA F2,303=25.223, p=<0.001), which were indistinguishable by δ15N (means
11.04‰ and 11.38‰ respectively).. The fish T. bernacchii and amphipod Heterophoxus
videns, had the highest δ15N values (13.98‰ and 13.76‰, respectively) although
samples of H. videns were somewhat variable (Figure 3). The δ15N signatures of T.
bernacchii were 7.15‰ more enriched than L. elliptica (trophic baseline), indicating
that four trophic levels exist within the Vestfold Hills food web.
126
Chapter V
Figure 2. δ13C values (mean ± SE) of fauna delineated by trophic category. Dashed line represents
pelagic/benthic carbon component offset by 0.1‰ per trophic level.
Predator/scavenger,
= Omnivore,
= Deposit feeder,
Carbon source. Gymnogrus sp. = -30.1‰ (not shown).
127
= Grazer,
= Predator,
=
= Suspension feeder,
=
Chapter V
Figure 3. δ15N values (mean ± SE) of fauna delineated by trophic category.
Predator/scavenger,
= Omnivore,
= Deposit feeder,
Carbon source. TL= trophic level
128
= Grazer,
= Predator,
=
= Suspension feeder,
=
Chapter V
Comparison with the Windmill Islands food web
With the exception of H. grandifolius and Chaetomorpha sp., δ13C values among
carbon sources were relatively similar for both regions (Figure 4a). Neither region
displayed consistent δ13C enrichment or depletion over the other location among carbon
sources (Figure 4a). However, δ13C values were generally enriched amongst the
Windmill Island fauna compared to those from the Vestfold Hills (Table 2, Figure 4b).
Vestfold Hills carbon sources were consistently enriched in δ15N compared to the same
sources from the Windmill Islands, with a mean enrichment of 2‰ amongst producers
between regions (range 0.10‰ to 3.8‰) (Table 2, Figure 5a). This trend continued
amongst consumers, with δ15N values consistently enriched by an average of 1.42‰
(range 0.43‰ to 5.75‰) (Table 2, Figure 5b).
Several notable similarities appeared between the Vestfold Hills food web and the
isotope food web model developed for the Windmill Islands (Figures 6a, 6b). Firstly,
sea ice POM was substantially enriched compared to all other carbon sources and
enrichment amongst POM carbon sources follows the trend: sea ice POM > sediment
POM > pelagic POM. Secondly, the suspension-feeding, deposit-feeding, and grazer
guilds occupied similar relative positions along the δ13C and δ15N planes. Lastly,
predators and scavenger/predators occupied a wide range of δ13C values, spanning the
δ13C range of lower-order consumers, but share similar positions in the δ15N plane with
predators only slightly, although significantly, enriched over predator/scavengers and
omnivores.
129
Chapter V
Figure 4. δ13C comparisons of (A) carbon sources and (B) consumers from the Vestfold Hills and
Windmill Islands.
Figure 5. δ15N comparisons of (A) carbon sources and (B) consumers from the Vestfold Hills and
Windmill Islands.
130
Chapter V
Table 2. Results of statistical comparisons of δ15N and δ13C isotope signatures of primary sources and
consumers collected from both the Vestfold Hills and Windmill Islands.
Samples
Sample size
p values for
comparisons of
15
δ N signatures
from Vestfold
Hills and
Windmill Islands
p values for
comparisons of
13
δ C signatures
from Vestfold
Hills and
Windmill Islands
Vestfold
Hills
Windmill
Islands
Chaetomorpha sp.
6
4
0.755
0.010*
Monostroma sp.
6
11
0.19
0.476
Himantothallus
6
9
<0.001*
<0.001*
Iridaea cordata
18
18
<0.001*
0.566
Palmaria decipiens
6
22
<0.001*
0.14
Pelagic POM
14
8
<0.001*
0.142
Sea Ice POM
47
17
0.467
0.354
Sediment POM
19
24
<0.001*
0.332
Adamussium colbecki
2
8
0.037*
0.001*
Cymodocella tubicauda
2
6
0.017*
0.022
Diplasterias brucei
11
11
<0.001*
0.001*
Heterophoxus videns
2
10
0.003*
0.057
Homaxinella sp.
5
4
0.503
0.126
Laternula elliptica
8
21
0.01*
<0.001*
Neobuccinium eatoni
18
12
0.009*
<0.001*
Nephtyidae
3
5
0.885
0.465
Nymphon australe
8
2
0.293
0.652
Odontaster validus
9
2
0.001*
0.214
Ophiura crassa
3
11
0.008*
0.097
Carbon source
grandifolius
Consumers
131
Chapter V
Paramoera walkeri
4
23
<0.001*
0.051
Parborlasia corrugatus
5
9
0.002*
0.023*
Perkinsiana sp.
8
36
0.773
<0.001*
Sterechinus neumayeri
12
21
<0.001*
<0.001*
Trematomus bernacchii
131
64
<0.001*
0.184
Urticinopsis antarctica
11
14
0.008*
0.241
Waldeckia obesa
3
4
0.004*
0.884
132
Chapter V
A
B
Figure 6. (A) Stable isotope values of carbon (δ13C) and nitrogen (δ15N) in carbon sources and consumers from the Vestfold Hills, and (B) stable isotope model of the Windmill
islands (reproduced from Gillies et al. 2012b)
133
Chapter V
Discussion
Vestfold Hills food web
We were successfully able to distinguish the main carbon sources and feeding guilds of
the Vestfold Hills food web through the use of stable isotope analysis. Lower-order
consumers could be grouped according to their feeding guild and main food sources
whilst higher-order consumers occupied the full range of δ13C ratios, indicating energy
at higher tropic levels originates from both benthic and pelagic carbon sources.
Omnivores were restricted towards the enriched end of the δ13C axis, indicating a
distinct origin of benthic food sources. There was little δ15N trophic partitioning
amongst predators, scavenger/predators and omnivores although higher-order guilds
were more enriched in δ15N compared to lower-order guilds, consistent with their
higher feeding status.
Carbon flow in the Vestfold Hills benthic community
The δ13C signatures of suspension feeders were closely aligned with the depleted δ13C
signatures in pelagic POM, confirming the majority of carbon utilised by suspension
feeders is derived from pelagic production. Yet there was considerable intra-guild
variation in δ13C (7‰) suggesting that other carbon sources are incorporated by at least
some suspension feeders. Dietary variation among Antarctic suspension feeders is not
unusual (Gili and Coma 1998, Gili et al. 2001) and is likely to be a result of extreme
temporal and spatial variability in primary food sources. For example, Norkko et al.
(2007) showed L. elliptica and A. colbecki can have variable diets, switching between
sea ice POM and detritus, in response to environmental gradients in food supply. Ahn
(1997) recorded that benthic diatoms, in addition to phytoplankton, constitute a
considerable portion of the diet of L. elliptica during summer, whilst smaller-sized
plankton fractions and resuspended sediment POM sustain some suspension feeders
during periods of limited food supply over winter (Barnes and Clarke 1995, Orejas et al.
2001).
Grazers and deposit feeders were clearly separated from suspension feeders on the δ13C
axis reflecting very different food sources. The δ13C values of grazers were, on average,
5-6‰ more enriched than pelagic POM demonstrating a diet based on benthic carbon
sources. However we found a mismatch between the δ13C signatures of deposit feeders
134
Chapter V
and grazers and the δ13C signatures of their (bulk) potential food sources. Gillies et al.
(2012b) found a similar mismatch between δ13C signatures in deposit feeders and
sediment POM, which may indicate Antarctic deposit feeders actively select different
carbon fractions based on their foraging location (Davenport 1988, BlazewiczPaszkowycz and Ligowski 2002, McMahon et al. 2006). For example, Cephalaspidea
and terrebellid polychaetes are likely to forage on or close to the sediment surface rather
than within the sediment and thus consume recently deposited sea ice algae,
phytoplankton detritus or benthic diatoms. Larger deposit feeders, such as O. crassa or
those that exclusively feed below the surface such as O. franklini, may consume less of
the fresher detritus or benthic diatoms and a greater proportion of older detritus which
has been enriched in δ13C as a consequence of bacterial processes (Macko et al. 1987,
Saino and Hattori 1987). Our results indicate that analysis of bulk sediment POM
samples does not provide enough resolution to determine specific diets for deposit
feeders, although δ13C signatures can clearly distinguish deposit feeders from other
first-order consumers. Future studies should seek to include gut content, DNA, or lipid
analysis, in combination with morphological analysis, to clearly determine diet in
deposit feeders.
We suspect several carbon sources not collected in this study account for the difference
in δ13C between grazers and macroalgae, with these sources likely occupying the δ13C
position between benthic macroalgae and sea ice POM. For example, Corbisier et al.
(2004) found benthic diatoms had δ13C signatures enriched by 7‰ over macroalgae and
9‰ over Pelagic POM, whilst Dunton (2001) found epiphytic algae had δ13C signatures
to be enriched by 2-6‰ over identical macroalgal species collected in this study.
Benthic diatoms and epiphytic algae have been shown to be an important food source in
other Antarctic shallow-waters regions (Iken 1999, Kaehler et al. 2000, Zacher et al.
2007), possibly due to the lack of chemical defences in comparison to most Antarctic
macroalgae (Amsler et al. 2005). We suggest they are equally likely to be important
components of the Vestfold Hills food web and advise future studies to ensure they are
represented as carbon sources.
Predator and scavenger/predator δ13C signatures spanned the carbon spectrum,
signifying higher trophic orders are ultimately reliant upon energy from several sources
derived from both benthic and pelagic environments. Our results are consistent with
other studies for Antarctic shallow water communities conducted on the Antarctic
135
Chapter V
Peninsula (Kaehler et al. 2000, Dunton 2001, Corbisier et al. 2004) and the Ross Sea
(Norkko et al. 2007) which found that consumers occupy similar trophic levels but
obtain their carbon from different sources. This is in contrast to benthic communities
occupying the Antarctic shelf, where productivity is largely driven by sinking
phytoplankton detritus (Jarre-Teichmann et al. 1997, Mincks et al. 2008).
Trophic structure
The δ15N values of predators, scavenger/predators and omnivores were similar,
indicating little trophic separation amongst these guilds, but they were enriched over
suspension feeders, deposit feeders and grazers by approximately 3.0‰, or one trophic
level. The fish T. bernacchii and amphipod Heterophoxus videns were enriched over all
other benthic predators by approx. 2.5‰, signifying the benthic food web occupies a
trophic range of four levels. However, we noted an almost continuous range of δ15N
ratios amongst the fauna. The lack of any distinct grouping along the δ15N axis suggests
a trophic continua (France et al. 1998) rather than traditional stepwise trophic models
(Hairston et al. 1960, Hairston and Hairston 1993). Trophic continua have previously
been demonstrated in shallow water communities from the Antarctic Peninsula (Kaehler
et al. 2000, Corbisier et al. 2004, Jacob et al. 2006) East Antarctica (Gillies et al. 2012b)
and on the continental shelf (Mincks et al. 2008) suggesting few guilds are tightly
coupled with specific food sources and omnivore (within feeding guilds) is common
amongst the benthos.
Based on these results, the exclusive use of trophic compartments to summarise food
web dynamics would misrepresent true feeding diversity and artificially reduce food
web connectance. For example, several Antarctic holothurians and cnidarians include a
considerable quantity of zooplankton in their diets (McClintock 1994, Orejas et al.
2001), which likely accounts for the elevated δ15N values we found in Cucumaria sp.,
Staurocucumis sp. and Pennatulacea. While the disadvantages of taxonomic or trophic
grouping in food webs are well documented (Cohen 1978, Polis 1991), the lack of
taxonomic certainty for many Antarctic fauna, particularly amongst lower-order
consumers, has probably led to considerable taxonomic or trophic grouping in previous
Antarctic food webs studies. Although we were unable to identify several lower-order
taxa to species level, our use of morphospecies identification highlights the considerable
variation amongst related suspension-feeders that would have been overlooked had
136
Chapter V
these species just been grouped at a higher taxonomic level. This suggests future studies
should look to identify species as morphospecies if they are unable to adequately
identify species with taxonomic certainty.
The variation in δ15N amongst primary consumers highlights the need for an isotopic
baseline, which allows the relative isotopic position of trophic levels to be standardised.
Several primary consumers had δ15N values below that of our isotopic baseline (L.
elliptica) indicating that δ15N values amongst carbon sources are inherently variable and
that the values in primary consumers reflect this variation. L. elliptica is known to feed
on a variety of organic materials (Ahn 1997, Norkko et al. 2007) confirming its position
as a first-order consumer. Those species which have isotopic values below that of L.
elliptica must, therefore, also be first order consumers, yet which feed on organic matter
that has lower δ15N values.
Vestfold Hills and Windmill Islands food webs
Despite variable δ13C amongst some carbon sources, the food webs for the Vestfold
Hills and Windmill Islands displayed surprisingly similar characteristics. We found
similar (within 1.5‰) δ13C signatures for pelagic POM, sea ice POM, sediment POM
and two common red algae (Iridaea cordata, Palmaria decipiens) and, to a lesser
extent, for the common brown alga Himantothallus grandifolius (2.5‰). This led to a
consistent scale of enrichment in δ13C amongst the main carbon sources in both food
webs (i.e. sea ice POM > sediment POM > pelagic POM) which is in agreement with
other polar studies (Hobson and Welch 1992, Hobson et al. 2002, Corbisier et al. 2004).
However, δ15N signatures in carbon sources were consistently enriched at Vestfold Hills
compared to the Windmill Islands. This continued in consumers, despite time-averaging
and trophic enrichment associated with temporal variability and trophic fractionation in
consumer tissue. These results imply source nitrogen in the Vestfold Hills food web is
enriched in δ15N compared to source nitrogen at the Windmill Islands resulting in a food
web-wide incremental shift in δ15N. Similar results have been found for comparable
food webs affected by sewage (Hadwen and Arthington 2007).
Enriched δ15N values are largely the result of inorganic source nitrogen assimilation in
producers, whereby 14N is preferentially selected over 15N during photosynthesis, when
under non-limiting conditions (Michener and Schell 1994). The δ15N signatures of
available inorganic source nitrogen may also vary spatially, and for marine systems,
137
Chapter V
source nitrogen from deep waters is generally enriched compared to surface waters
(Conway et al. 1994, Cabana and Rasmussen 1996). Waters located close to upwelling
zones may consequently have a sustained source of enriched nitrogen compared to areas
further away.
Although both the Vestfold Hills and Windmill islands share similar local
characteristics (e.g. prevalence of rocky islands close to the coast, several shallow bays)
there are considerable oceanographic differences which may account for variation in
source nitrogen between the two locations. The Vestfold Hills lie on the eastern side of
Prydz Bay, the third largest embayment on the Antarctic coast. Horizontal currents
within the bay are dominated by a large, clock-wise flowing gyre, connecting shelf and
continental waters with the offshore circum-Antarctic current and a narrow, coastal
current that flows west. The Vestfold Hills are located near the entry point of the gyre
and coastal current and hence, are likely to receive deep, nutrient rich waters (Smith and
Treguer 1994, Nunes Vaz and Lennon 1996). Local upwelling events in the vicinity of
the Vestfold Hills can occur (Anilkumar et al. 2010), which may further increase the
flow of deep water into shallow areas. In contrast, the Windmill Islands lie on the
exposed coast and currents would be expected to vary considerably (i.e. lack of gyres
and large ocean upwelling) compared to those within Prydz Bay. Unfortunately, the
lack of physical oceanographic data from the Windmill Islands precludes further
exploration of these differences. Nevertheless, our results highlight the effect of large
scale oceanographic processes on food webs represented by stable isotopes.
The relative positioning of each feeding guild on the isotopic bi-plots is similar for both
food webs indicating stable isotopes were able to summarise similar aspects of carbon
flow and trophic position across large spatial scales. In both food webs, suspension
feeders occupied a similar position with depleted δ13C and δ15N signatures, whilst
grazers occupied the same trophic position but were enriched in δ13C corresponding
with differences in diet.
These consistencies demonstrate that isotope-derived food webs describing similar
communities are able to display consistent properties facilitating Antarctic-wide
generalisations about community structure and function. The use of stable isotopes to
compare food webs is still in its infancy (Layman et al. 2007, Hoeinghaus and Zeug
2008, Layman and Post 2008) yet shows promising signs of contributing to important
ecological concepts such as niche width (Bearhop et al. 2004, Newsome et al. 2007) and
138
Chapter V
ecosystem functioning (Vander Zanden and Fetzer 2007). Our study shows that, for at
least two nearshore communities separated by over 1000 km in East Antarctica, stable
isotopes are a useful tool in measuring food web similarity.
Conclusion
The benthic community of the Vestfold Hills obtains its energy from several carbon
sources from both pelagic and benthic origins and displayed similar characteristics to
that described for the Windmill Islands. Further analysis with studies conducted from
other regions in Antarctica, would prove fruitful in refining a representative food web
model for Antarctic, high-latitude, shallow water ecosystems. Such a model would
provide a trophic benchmark against which modification in these communities brought
about by climate change or other human impacts could be compared.
139
Chapter V
Acknowledgments
We are grateful to members of 2009/10 summer Davis field and diving teams for
assistance with field collections. Advice on stable isotope preparation was provided by
Hillary Stuart-Williams. Figure 1 was produced by Kathryn James from maps provided
by the Australian Antarctic Data Centre. This research was funded by a PhD research
scholarship from Southern Cross University and supported financially and logistically
by the Australian Antarctic Division (AAS projects 2948 and 2201).
140
Chapter V
References
Ahn I (1997) Feeding ecology of the Antarctic lamellibranch Laturnula elliptica
(Laturnulidae) in Marian Cove and vicinity King George Island, during one austral
summer. In: Battaglia B, Valencia J, Walton DWH (eds) Antarctic communities:
species structure and survival. Cambridge University Press, Cambridge
Amsler CD, McClintock JB, Baker BJ (1999) An Antarctic feeding triangle: defensive
interactions between macroalgae, sea urchins, and sea anemones. Mar Ecol Prog Ser
183:105-114
Amsler CD, Iken K, McClintock J, Amsler MO, Peters KJ, Hubbard JM, Furrow FB,
Baker B (2005) Comprehensive evaluation of the palatability and chemical defences of
subtidal macroalgae from the Antarctic Peninsula. Mar Ecol Prog Ser 294:141-159
Anderson, M.J., 2001. Permutation tests for univariate or multivariate analysis of
variance and regression. Can J Fish Aquat Sci 58, 626-639.
Anilkumar N, Mohan R, Shukla SK, Pednekar S, Sudhakar M, Ravindra R (2010)
Signature of coastal upwelling in Prydz Bay, East Antarctica during austral summer
2006. Curr Sci 99:1390-1394
Arntz W, Brey T, Gallardo VA (1994) Antarctic zoobenthos. Oceanogr Mar Biol Annu
Rev 32:241-304
Barbraud C, Weimerskirch H (2001) Emperor penguins and climate change. Nature
411:183-186
Barnes DKA, Clarke A (1995) Seasonality of feeding activity in Antarctic suspension
feeders. Polar Biol 15:335-340
Barnes D, Linse K, Waller C, Morely S, Enderlein P, Fraser K, Brown M (2006)
Shallow benthic fauna communities of South Georgia Island. Polar Biol 29:223-228
Barnes DKA (1999) The influence of ice on the polar nearshore benthos. J Mar Biol
Assoc UK 79:401-407
Bearhop S, Adams CE, Waldron S, Fuller RA, Macleod H (2004) Determining trophic
niche width: a novel approach using stable isotope analysis. J Anim Ecol 73:1007-1012
Blazewicz-Paszkowycz M, Ligowski R (2002) Diatoms as food source indicator for
some Antarctic Cumacea and Tanaidacea (Crustacea). Antarct Sci 14:11-15
141
Chapter V
Cabana G, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen
isotopes. Proc Natl Acad Sci USA 93:10844-10847
Carabel S, Godinez-Dominguez E, Verisimo P, Fernandez L, Freire J (2006) An
assessment of sample processing methods for stable isotope analyses of marine food
webs. J Exp Mar Biol Ecol 336:254-261
Casaux R, Barrera-Oro E, Baroni A, Ramón A (2003) Ecology of inshore notothenioid
fish from the Danco Coast, Antarctic Peninsula. Polar Biol 26:157-165
Clarke A (1988) Seasonality in the Antarctic marine environment. Comp Biochem
Physiol B 90:461-473
Conway NM, Kennicutt I MC, Van Dover CL (1994) Stable isotopes in the study of
marine chemosynthetic-based ecosystems. In: Lajtha K, Michener R (eds) Stable
isotopes in ecology and environmental science. Blackwell scientific publications,
London
Corbisier T, Petti MV, Skowronski RP, Brito TS (2004) Trophic relationships in the
nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C stable-isotope
analysis. Polar Biol 27:75-82
Dauby P, Scailteur Y, De Broyer C (2001) Trophic diversity within the eastern Weddell
Sea amphipod community. Hydrobiolo 443:69-86
Davenport J (1988) The feeding mechanism of Yoldia (= Aequiyoldia) eightsi
(Courthouy). Proc R Soc Lond, Ser B: Biol Sci 232:431-442
Dayton PK, Gordon AR, Paine RT, Dayton LB (1974) Biological accommodation in the
benthic community at McMurdo Sound, Antarctica. Ecol Monogr 44:105-128
Dayton PK, Robilliard GA, Paine RT (1970) Benthic faunal zonation as a result of
Anchor Ice at McMurdo Sound, Antarctica. In: Holdgate M (ed) Antarctic Ecology,
Book 1. Academic Press, London
Dayton PK, Oliver JS (1977) Antarctic soft-bottom benthos in oligotrophic and
eutrophic environments. Science 197:55-58
Dayton PK, Mordida BJ, Bacon F (1994) Polar marine communities. Am Zool 34:90-99
Dearborn JH (1967) Food and reproduction of Glyptonotus antarcticus (Crustacea:
Isopoda) at McMurdo Sound, Antarctica. T Roy Soc NZ 8:163-168
142
Chapter V
Dearborn JH, Hendler G, Edwards KC (1996) The diet of Ophiosparte gigas
(Echinodermata: Ophiuroidea) along the Antarctic Peninsula, with comments on its
taxonomic status. Polar Biol 16:309-320
Deniro MJ, Epstein S (1977) Mechanism of carbon isotope fractionation associated with
lipid synthesis. Science 197:261-263
Dhargalkar VK (1990) Benthic marine algae of the inshore water at the Vestfold Hills,
Antarctica. Indian J Mar Sci 19:110-114
Dunton KH (2001) δ15N and δ13C measurements of Antarctic Peninsula fauna: trophic
relationships and assimilation of benthic seaweeds. Amer Zool 41:99-112
Everitt D, Poore G, Pickard J (1980) Marine benthos from Davis Station, Eastern
Australia. Mar Freshw Res 31:829-836
Fauchald K, Jumars PA (1979) The diet of worms: a study of polychaete feeding guilds.
Oceanogr Mar Biol Annu Rev 17:193-284
France RL, Chandler M, Peters R (1998) Mapping trophic continua of benthic
foodwebs: body size-δ15N relationships. Mar Ecol Prog Ser 174:301-306
Gambi MC, Lorenti M, Russo GF, Scipione MB (1994) Benthic associations of the
shallow hard bottoms off Terra Nova Bay, Ross Sea: zonation, biomass and population
structure. Antarct Sci 6:449-462
Gili J-M, Coma R (1998) Benthic suspension feeders: their paramount role in littoral
marine food webs. Trends Ecol Evol 13:316-321
Gili J-M, Coma R, Orejas C, López-González P, Zabala M (2001) Are Antarctic
suspension-feeding communities different from those elsewhere in the world? Polar
Biol 24:473-485
Gillies C, Stark J, Smith S (2012a) Small-scale spatial variation of δ13C and δ15N
isotopes in Antarctic carbon sources and consumers. Polar Biol 35:813-827
Gillies CL, Stark JS, Johnstone GJ, Smith SDA (2012b) Carbon flow and trophic
structure of an Antarctic coastal benthic community as determined by δ13C and δ15N.
Estuar Coast Shelf Sci 97:44-57
Gutt J (1991) On the distribution and ecology of holothurians in the Weddell Sea
(Antarctica). Polar Biol 11:145-155
143
Chapter V
Gutt J (2007) Antarctic macro-zoobenthic communities: a review and an ecological
classification. Antarct Sci 19:165-182
Hadwen WL, Arthington AH (2007) Food webs of two intermittently open estuaries
receiving 15N-enriched sewage effluent. Estuar Coast Shelf Sci 71:347-358
Hairston NG, Jr., Hairston NG, Sr. (1993) Cause-effect relationships in energy flow,
trophic structure, and interspecific interactions. Amer Nat 142:379-411
Hairston NG, Smith FE, Slobodkin LB (1960) Community structure, population control,
and competition. Amer Nat 94:421-425
Hobson KA, Fisk A, Karnovsky N, Holst M, Gagnon J-M, Fortier M (2002) A stable
isotope (δ13C, δ15N) model for the North Water food web: implications for evaluating
trophodynamics and the flow of energy and contaminants. Deep Sea Res Pt II 49:51315150
Hobson KA, Welch HE (1992) Determination of trophic relationships within a high
Arctic marine food web using δ13C and δ15N analysis. Mar Ecol Prog Ser 84:9-18
Hoeinghaus DJ, Zeug SC (2008) Can stable isotope ratios provide for community-wide
measures of trophic structure? Comment. Ecology 89:2353-2357
Iken K (1999) Feeding ecology of the Antarctic herbivorous gastropod Laevilacunaria
antarctica Martens. J Exp Mar Biol Ecol 236:133-148
Jacob U, Brey T, Fetzer I, Kaehler S, Mintenbeck K, Dunton K, Beyer K, Struck U,
Pakhomov EA, Arntz WE (2006) Towards the trophic structure of the Bouvet Island
marine ecosystem. Polar Biol 29:106-113
Jacob U, Mintenbeck K, Brey T, Knust R, Beyer K (2005) Stable isotope food web
studies: a case for standardized sample treatment. Mar Ecol Prog Ser 287:251-253
Jarre-Teichmann A, Brey T, Bathman U, Dahm C, Dieckmann G, Gorny M, Klages M,
Pages F, Plotz J, Schnack-Schiel SB, Stiller M, Arntz W (1997) Trophic flows in the
benthic shelf community of the eastern Weddell Sea, Antarctica. In: Antarctic
communities: species, structure and survival. Cambridge University Press, Cambridge
Johnston E, Connell S, Irving A, Pile A, Gillanders B (2007) Antarctic patterns of
shallow subtidal habitat and inhabitants in Wilke’s Land. Polar Biol 30:781-788
144
Chapter V
Kaehler S, Pakhomov EA, Mcquaid CD (2000) Trophic structure of the marine food
web at the Prince Edward Island (Southern Ocean) determined by 13C and 15N analysis.
Mar Ecol Prog Ser 208:13-20
Kiest KA (1993) A relationship of diet to prey abundance and the foraging behaviour of
Trematomus bernacchii. Polar Biol 13:291-296
La Mesa M, Dalú M, Vacchi M (2004) Trophic ecology of the emerald notothen
Trematomus bernacchii (Pisces, Nototheniidae) from Terra Nova Bay, Ross Sea,
Antarctica. Polar Biol 27:721-728
Lane A, Riddle MJ (2004) Toxicity of dispersed and undispersed diesel in sediments to
the Antarctic Ophiuroid Ophiura crassa. Australas J Ecotoxicol 10:115-121
Layman C, Arrington D, Montaña C, Post D (2007) Can stable isotope ratios provide
for community-wide measures of trophic structure? Ecology 88:42-48
Layman CA, Post DM (2008) Can stable isotope ratios provide for community-wide
measures of trophic structure? Reply. Ecology 89:2358-2359
Macdonald RW, Harner T, Fyfe J (2005) Recent climate change in the Arctic and its
impact on contaminant pathways and interpretation of temporal trend data. Sci Total
Environ 342:5-86
Macko SA, Fogel ML, Hare PE, Hoering TC (1987) Isotopic fractionation of nitrogen
and carbon in the synthesis of amino acids by microorganisms. Chem Geolo 65:79-92
Massom RA, Stammerjohn SE (2010) Antarctic sea ice change and variability Physical and ecological implications. Polar Sci 4:149-186
McArdle, B.H., Anderson, M.J., 2001. Fitting multivariate models to community data: a
comment on distance based redundancy analysis. Ecology 82, 290-297.
McCann KS (2000) The diversity-stability debate. Nature 405:228-233
McClintock JB (1994) Trophic biology of Antarctic shallow-water echinoderms. Mar
Ecol Prog Ser 111:191-202
McConnaughey T, McRoy CP (1979) Food-web structure and the fractionation of
carbon isotopes in the Bering Sea. Mar Biol 53:257-262
McCutchan JH, Lewis WM, Kendall C, McGrath CC (2003) Variation in trophic shift
for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102:378-390
145
Chapter V
McMahon KW, Ambrose Jr WG, Johnson BJ, Sun M-Y, Lopez GR, Clough LM,
Carroll ML (2006) Benthic community response to ice algae and phytoplankton in Ny
Ålesund, Svalbard. Mar Ecol Prog Ser 310:1-14
Michener R, Schell DM (1994) Stable isotopes as tracers in marine aquatic food webs.
In: Lajtha K, Michener R (eds) Stable isotopes in ecology and environmental science.
Blackwell scientific publications, London
Mincks SL, Smith CR, Jeffreys RM, Sumida PYG (2008) Trophic structure on the West
Antarctic Peninsula shelf: detritivory and benthic inertia revealed by δ13C and δ15N
analysis. Deep Sea Res Pt II 55:2502-2514
Montgomery JC, Foster BA, Milton RC, Carr E (1993) Spatial and temporal variations
in the diet of nototheniid fish in McMurdo Sound, Antarctica. Polar Biol 13:429-431
Moran MD (2003) Arguments for rejecting the sequential Bonferroni in ecological
studies. Oikos 100:403-5
Murray MD, Luders DJ (1990) Faunistic studies at the Windmill Islands, Wilkes Land,
East Antarctica, 1959-80. ANARE Research Notes 73:45pp
Newsome S, Martinez Del Rio C, Bearhop S, Phillips D (2007) A niche for isotopic
ecology. Front Ecol Environ 5:429-436
Nonato EF, Brito TAS, De Paiva PC, Petti MAV, Corbisier TN (2000) Benthic
megafauna of the nearshore zone of Martel Inlet (King George Island, South Shetland
Islands, Antarctica): depth zonation and underwater observations. Polar Biol 23:580588
Norkko A, Thrush SF, Cummings V, Gibbs HL, Andrew N, Norkko J, Schwarz A-M
(2007) Trophic structure of coastal Antarctic food webs associated with changes in sea
ice and food supply. Ecology 88:2810-2820
Nunes Vaz RA, Lennon GW (1996) Physical oceanography of the Prydz Bay region of
Antarctic waters. Deep Sea Res Pt I 43:603-641
Nyssen F, Brey T, Lepoint G, Bouquegneau J-M, De Broyer C, Dauby P (2002) A
stable isotope approach to the eastern Weddell Sea trophic web: focus on benthic
amphipods. Polar Biol 25:280-287
146
Chapter V
Oliver JS, Slattery PN (1985) Effects of crustacean predators on species composition
and population structure of soft-bodied infauna from McMurdo Sound, Antarctica.
Ophelia 24:155-175
Orejas C, Gili J, López-González P, Arntz W (2001) Feeding strategies and diet
composition of four Antarctic cnidarian species. Polar Biol 24:620-627
Pearse JS, Giese AC (1966) Food, reproduction and organic constitution of the
Antarctic echinoid Sterechinus neumayeri (Meissner). Biolo Bull 130:387-401
Peck L (2005) Prospects for surviving climate change in Antarctic aquatic species.
Front Zoolo 2:9
Polis GA (1994) Food webs, trophic cascades and community structure. Aust J Ecol
19:121-136
Post D, Layman C, Arrington D, Takimoto G, Quattrochi J, Montaña C (2007) Getting
to the fat of the matter: models, methods and assumptions for dealing with lipids in
stable isotope analyses. Oecologia 152:179-189
Post DM (2002a) The long and short of food-chain length. Trends Ecol Evol 17:269277
Post DM (2002b) Using stable isotopes to estimate trophic position: models, methods
and assumptions. Ecology 83:703-718
Purinton BL, Demaster DJ, Thomas CJ, Smith CR (2008)
14
C as a tracer of labile
organic matter in Antarctic benthic food webs. Deep Sea Res II 55:2438-2450
Quartino M, Boraso De Zaixso A (2008) Summer macroalgal biomass in Potter Cove,
South Shetland Islands, Antarctica: its production and flux to the ecosystem. Polar Biol
31:281-294
Quinn G, Keough M (2002) Experimental design and data analysis for biologists, Vol.
Cambridge University Press, Cambridge
Rakusa-Suszczewski S (1972) The biology of Paramoera walkerii Stebbing
(Amphipoda) and the Antarctic sub-fast ice community Pol Arch Hydrobiolo 19:11-36
Saino T, Hattori A (1987) Geographical variation of the water column distribution of
suspended particulate organic nitrogen and its δ15N natural abundance in the Pacific and
its marginal seas. Deep Sea Res Pt I 34:807-827
147
Chapter V
Smale DA, Barnes DKA, Fraser KPP, Mann PJ, Brown MP (2007) Scavenging in
Antarctica: Intense variation between sites and seasons in shallow benthic necrophagy. J
Exp Mar Biol Ecol
Smith N, Treguer P (1994) Physical and chemical oceanography in the vicinity of Prydz
Bay, Antarctica. In: El-Sayed Z (ed) Southern Ocean ecology: the BIOMASS
perspective. Cambridge University Press, Cambridge
Snape I, Riddle MJ, Stark JS, King CK, Duquesne S, Gore DB (2001) Management and
remediation of contaminated sites at Casey Station, Antarctica. Polar Rec 37:199-214
Stammerjohn SE, Martinson DG, Smith RC, Iannuzzi RA (2008) Sea ice in the western
Antarctic Peninsula region: spatio-temporal variability from ecological and climate
change perspectives. Deep Sea Res Pt II 55:2041-2058
Tucker MJ, Burton HR (1987) A survey of the marine fauna in shallow coastal waters
of the Vestfold Hills and Rauer Islands, Antarctica. ANARE Research Notes 55:1-27
Tucker MJ, Burton HR (1988) The inshore marine ecosystem off the Vestfold Hills,
Antarctica. Hydrobiologia 165:129-139
Vander Zanden MJ, Fetzer WW (2007) Global patterns of aquatic food chain length.
Oikos 116:1378-1388
Vander Zanden MJ, Rasmussen JB (2001) Variation in δ15N and δ13C trophic
fractionation: Implications for aquatic food web studies. Limnol Oceanogr 46:20612066
Vanderklift M, Ponsard S (2003) Sources of variation in consumer-diet δ15N
enrichment: a meta-analysis. Oecologia 136:169-182
Wägele H (1989) Diet of some Antarctic nudibranchs (Gastropoda, Opisthobranchia,
Nudibranchia). Mar Biol 100:439-441
Williams R (1988) The inshore marine fishes of the Vestfold Hills region, Antarctica.
Hydrobiolo 165:161-167
Zacher K, Hanelt D, Wiencke C, Wulff A (2007) Grazing and UV radiation effects on
an Antarctic intertidal microalgal assemblage: a long-term field study. Polar Biol
30:1203-1212
148
Chapter V
Appendix
Appendix 1. Authors for feeding guilds.
Taxonomic
group
Authors for feeding guilds
Amphipoda
Dauby et al. (2001), Oliver and Slattery (1985) Rakusa-Suszczewski (1972)
Nyssen et al. (2002)
Anthoza
Ascidiacea
Asteroidea
McClintock (1994)
Bivalvia
Ahn (1997), Norkko et al. (2007)
Cnidaria
Amsler et al (1999), Dayton et al. (1970), This study
Decapoda
This study
Echinoidea
McClintock (1994) Norkko et al. (2007), Pearse and Giese (1966)
Gastropoda
Norkko et al. (2007), Dayton et al. (1970), This study, Wagele (1989)
Holothuriodea
Gutt (1991), McClintock (1994) Gillies et al. (2012b)
Hydroza
Isopoda
Dearborn (1967) Gillies et al. (2012b)
Nemertia
Corbisier et al. (2004) Smale et al. (2007)
Ophuroidea
Dearborn et al. (1996), McClintock (1994), Lane and Riddle (2004)
Perciformes
Kiest (1993), La Mesa et al. (2004) Casaux et al. (1990), Casaux et al.
(2003), Foster and Montgomery (1993), Williams (1988)
Polychaeta
Fauchald and Jumars (1979)
Porifera
Pycnognida
This study, Gillies et al. (2012b)
Tanaidacia
Oliver and Slattery (1985), Blazewicz-Paszkowycz and Ligowski (2002)
Gillies et al. (2012b)
149
Chapter V
150
Chapter VI
Chapter VI
A synthesis of stable isotope food web studies from marine
systems: what can they contribute to food web theory?
Christopher L. Gillies, Jonathan S. Stark and Stephen D. A. Smith
In preparation for submission to Oikos or Marine Biology
151
Chapter VI
Abstract
Determining which forces control the structure of communities relies on testing theories
with empirical evidence through comparative analysis amongst different systems.
Despite the common application of stable isotopes to describe carbon flow and trophic
structure in studies of single food webs, the capacity for comparative analysis among
food web studies utilising stable isotopes (referred to as stable isotope food web studiesSIFWS) to contribute to food web theory has not been critically assessed. In this study
we evaluated 97 marine SIFWS across several different habitats and compared common
food web metrics to assess the validity of SIFWS data for cross-system comparisons
and to detect large-scale patterns to inform food web theory. We found data from
SIFWS could be categorised according to basic ecological and community descriptors
such as community size, resource availability and habitat. Food chain length was similar
across all habitats, averaging close to four trophic levels, and was weakly, though
significantly, negatively correlated with mean oceanographic primary productivity, and
positively correlated with top δ15N (maximum δ15N among species). Carbon (δ13C)
range in consumers was found to be a useful indication of resource availability and was
linked to the number of producer groups sampled. Pelagic food webs had narrower
carbon ranges compared to benthic food webs, or food webs which contained both
habitats, a result consistent with the reduced number of producer groups in pelagic
habitats. We highlight the current limitations of comparative analysis across SIFWS,
due largely to the lack of consistent reporting of sampling effort and clear delineation of
food web boundaries. Despite these limitations, this study demonstrates comparative
analysis of SIFWS can identify broad-scale patterns in food web metrics relevant to
food web theory.
Key words: trophic, food chain length, carbon range, niche, comparative analysis
152
Chapter VI
Introduction
Since the early 1970s, stable isotopes of carbon and nitrogen have commonly been used
by ecologists to depict carbon flow and trophic structure in food web studies (Peterson
1999, Post 2002b). Examination of the literature, however, reveals that food web
studies that have been developed using stable isotopes (hereafter referred to as stable
isotope food web studies - SIFWS) are overwhelmingly dominated by studies conducted
in a single location or among few systems. Comparative or synthesis studies among
multiple systems or habitats are rare, despite being a critical component in elucidating
the mechanisms underlying ecosystem function and in the testing of ecological theories
(Schoener 1989). Although there have been regular reviews identifying the strengths
and weaknesses of isotopic analysis as a standard tool to elucidate dietary and trophic
links amongst species within food webs (Wada et al. 1991, Peterson 1999, Pasquaud et
al. 2007 and references there in), there has been little evaluation of the relevance of
SIFWS as a body of literature that could be analysed to detect global patterns amongst
community structure and function (Post 2002b, Vander Zanden and Fetzer 2007). Such
analyses would be extremely useful in empirical testing and further advancement of
ecological theories.
Food webs studies derived from traditional methods such as field observation, gut
content or feeding analysis have been criticised for their over-simplification, lack of
standardisation and variable sampling effort (Paine 1988, Closs 1991, Polis 1991). Each
of these limits the comparative analysis of food webs across different systems. Hence,
major elements of food web theory, such as factors governing food chain length,
community complexity and stability, have considerably relied upon theoretical
modelling and testing (May 1973, Dunne et al. 2005). Despite the advances made by
theoretical models, ecological theories, especially those relating to food webs, require
considerable empirical testing before they can become widely accepted (Paine 1988).
Thus, it is essential for ecologists to develop methods which facilitate comparative
analysis amongst real food webs.
SIFWS have several promising features which may be advantageous over traditional
(conceptual) connectivity webs in terms of standardising sampling and analytical
methodologies to facilitate cross-system comparisons. For example, by reducing the
spatial and temporal isotopic variation associated with fast-growing primary producers,
153
Chapter VI
isotopic baselines provide a method for standardising trophic position at the base of the
food chain (Vander Zanden and Rasmussen 1999, Post 2002b). This approach facilitates
comparative analysis among different systems (Cabana and Rasmussen 1996, Vander
Zanden and Fetzer 2007). The development of mixing models, that can partition more
than two food sources in a consumer’s diet (Phillips et al. 2005, Moore and Semmens
2008), and the use of compound-specific isotopic analysis of amino acids to more
accurately determine diet and trophic position (Fantle et al. 1999, McClelland and
Montoya 2002, Chikaraishi et al. 2009), demonstrate a growing trend in increasing
isotopic detail. This suggests that, in the near future, isotopic food webs will soon
resemble connectance webs in the amount of dietary information they contain.
Furthermore, the recent increased application of stable isotopes has resulted in reduced
analytical costs, which should facilitate greater sample sizes, increased replication and
the assembly of global isotopic data sets (e.g. Raymond et al. 2011). Thus, it appears
that analyses of stable isotopes are close to surpassing traditional dietary techniques as
the method of choice in large-scale dietary studies.
The few studies that have utilised stable isotopes (or isotopic indices) comparatively
across systems have indicated strong potential to contribute to further understanding of
ecological concepts. For example, δ15N range (the distance between the lowest and
highest δ15N values within a food web) is an accurate measure of trophic height or food
chain length (FCL) (Carpenter et al. 1985, Post et al. 2000). FCL is regarded as a
fundamental parameter that can summarise community function and dynamics
(Oksanen et al. 1981, Fretwell 1987) and analysis of FCL can facilitate comparisons of
communities in different habitats or locations (Cabana and Rasmussen 1996, Post
2002a, Vander Zanden and Fetzer 2007). The δ15N ratios in consumers are an accurate
predictor of trophic position (Deniro and Epstein 1981, Post 2002b), which can be
correlated with body size (Akin and Winemiller 2008), a key metric in the
determination of interaction strength between species or in deciphering predator-prey
relationships (Jennings et al. 2001, Brose et al. 2006). The ‘isotopic niche’ of a species
may reflect its ecological niche (Bearhop et al. 2004, Newsome et al. 2007), whilst δ13C
range (the distance between the lowest and highest δ13C values within a food web) can
provide a measure of the available dietary niche space within a food web (Layman et al.
2007).
154
Chapter VI
Despite the promising nature of comparative analysis of SIFWS to identify large-scale
food web patterns, no study to date, has specifically addressed concerns relating to data
validity. Variable sampling effort amongst studies may inhibit cross-system
comparisons if food web studies are not representative of actual community structure.
This may occur if SIFWS fail to include a representative proportion of food sources and
consumers when describing food webs, obviating the detection of large-scale patterns.
Validating fundamental assumptions on SIFWS is, therefore, an essential ‘first step’ in
conducting large-scale, cross-system comparisons. This study sets out to address this
concern by examining the available marine SIFWS to compile data on food web
attributes and isotopic indices across habitats. Our primary goals are to: (1) assess the
state of data from SIFWS to determine if they conform to basic ecological descriptors;
and (2) identify large-scale patterns relating to food web theory.
We concentrate on two metrics important to food web theory: food chain length and
trophic niche width. We specifically calculated FCL and δ13C range from marine
SIFWS and compared these to several food web elements: community size, resource
availability and location (latitude, habitat and environment). We test the theory that
FCL varies according to primary productivity or habitat. In so doing, we review the
value and limitations of marine SIFWS and their ability to provide metrics which can be
standardised across all studies. We also review the usefulness of δ13C range (CR) as a
measure of dietary niche width (Layman et al. 2007).
155
Chapter VI
Methods
Stable isotope data from marine food webs were obtained from published studies found
through searching ISI Web of Knowledge using a combination of the terms:
‘isotope(s)’, ‘food web(s)’, ‘marine’ and ‘trophic’. We also examined the citations listed
in the source articles.
Food web indices
We chose two simple isotopic indices, FCL (δ15N range) and CR (δ13C range), both of
which show promise for cross-system comparisons and ease of calculation (e.g. Post et
al. 2000, Layman et al. 2007, Vander Zanden and Fetzer 2007). However, they have yet
to be sufficiently tested across multiple systems. We also assess the possibility of using
top δ15N (maximum δ15N among species) as a surrogate for FCL.
Food chain length (FCL)
We define FCL as δ15N range i.e. the distance between the most enriched and depleted
species within a food web. A larger range in δ15N among consumers suggests more
trophic levels and thus a greater degree of trophic diversity (Layman et al. 2007).
FCL for each food web were estimated using the formula:
FCL= (δ15Ntop predator - δ15Nbaseline) / 3.4 + λ
where 3.4 is the fractionation of δ15N between trophic levels and λ is the trophic level of
the baseline indicator, set at 2 due to the use of primary consumers as a baseline. This
method has been used previously (Post et al. 2000, Vander Zanden and Fetzer 2007)
and is applied here for consistency with these studies. Other studies have found lower
fractionation values in consumers with invertebrate diets, and for aquatic herbivores
(Vander Zanden and Rasmussen 2001, McCutchan et al. 2003). Slightly higher
fractionation rates (4.1‰) have been found for Antarctic consumers (Kaehler et al.
2000).
To avoid confounding of comparisons due to the inherent spatial and temporal variation
amongst producers (e.g. Gillies et al. 2012), the δ15N values of primary consumers were
used from each food web to estimate the trophic base, as primary consumers are better
able to integrate the temporal and spatial variation of autotrophs (Vander Zanden and
Rasmussen 1999, Post 2002b). Where possible, similar taxa were consistently used as
156
Chapter VI
baseline primary consumers (e.g. calanoid copepods, herbivorous gastropods, bivalves)
except when alternate baseline species had been previously identified by the study.
Previous bootstrap simulation conducted by Vander Zanden and Fretzer (2007)
indicates that the use of different primary producers to calculate baseline δ15N results in
a mean error variance of 0.29 (1 SD, n= 26) (for marine studies).
Our results for FCL differ slightly from the original estimates for some studies because
we used a universal fractionation value which ensures a consistent measure across all
food webs. We noted few studies sampled marine mammals, reptiles or birds: we
therefore omitted these groups from the analysis to avoid bias in those studies where
these groups were present but not sampled. We note Vander Zanden and Fretzer (2007)
took a similar approach to avoid bias in transient species. We acknowledge the open
nature of marine systems is a limiting yet important factor in deciphering food web
boundaries and suggest studies should comment, at minimum, on the presence or
absence of species which are not always present yet may affect food web dynamics.
We assess whether a higher FCL is represented by a higher top (i.e. maximum) δ15N. It
is assumed a higher FCL would, intuitively, result in a higher top δ15N: however, this
assumption has not been adequately assessed. FCL is dependent on both the δ15N of the
base carbon source and δ15N of the top predator. Therefore, a higher top δ15N may not
necessarily correlate with a high FCL, but simply reflect more enriched food sources in
some habitats or locations. Such a scenario would preclude the use of top δ15N as a
valid metric for cross-system comparisons.
δ13C range (CR)
We define the available dietary niche width within a food web as δ13C range i.e. the
distance between the two most enriched and depleted consumer species within a food
web, defined as CRconsumers. We also calculate the δ13C range of producers defined as
CRproducers i.e. the distance between the two most enriched and depleted base carbon
sources.
Layman et al. (2007) hypothesised that higher CRconsumers values can be expected in food
webs in which there are multiple carbon sources, providing for niche diversification at
the base of a food web. However, this hypothesis has come under scrutiny, as the use of
157
Chapter VI
CRconsumers as a measure of niche diversity may only be applicable in systems that share
a similar number of carbon sources with comparable isotopic signatures, because
variation in CRconsumers is a product of the isotopic variation amongst producers
(Hoeinghaus and Zeug 2008). We test this hypothesis: that CRconsumers is dependent on
the number of producers available with the Ho being CRconsumers is independent of the
number of producers. A rejection of our null hypothesis would result in CRconsumers
being a suitable measure of available niche diversity.
As a result of isotopic averaging with increasing trophic level, CRconsumers should be less
than CRproducers if a sufficient proportion of carbon sources were sampled. We test
whether this is consistent across food webs, as a CRconsumers > CRproducers would indicate
under-representation of basal food sources in the food web, thus considerably limiting
the interpretation of food web results.
Food web elements
2.2.1. Nomenclature
We grouped each study into one of three categories based on the environment of the
majority of species within each study: benthic, pelagic, or both (when studies did not
differentiate or included both benthic and pelagic species). In some instances, one or
two pelagic species were included in studies determined as ‘benthic’ and vice versa.
However, these species were rarely listed as either top predators or as limits for
calculating CR and, therefore, their inclusion should have minimal effect on estimates
of FCL or CR. We further divided benthic studies into three zones based on: their broad
ecosystem - estuarine, coastal, or shelf/deep sea; and as one of ten habitats based on site
descriptions - coastal sediment, rocky/sediment, mangroves, rocky/macroalgae, salt
marsh, salt marsh/mangroves, sea grass, seagrass/mangrove, sandy beach and shelf/deep
sea sediment.
Resource availability
We used productivity estimates as a measure of resource availability, to test the
‘productivity hypothesis’ which states that FCL increases with increasing productivity
as a result of greater energy being available for transfer between trophic levels (Elton
1927, Hutchinson 1959). To estimate marine productivity for each study, we used
158
Chapter VI
depth-integrated chlorophyll biomass (mgC/m2/day), averaged over the years 20032010, using the mean nearest 0.5° latitude and longitude for the location stated in the
original study. Values were obtained from SeaWiFS data available from the Oregon
State
University
Ocean
Productivity
website
(http://www.science.oregonstate.edu/ocean.productivity/ accessed 15/06/11). We also
used latitude as a proxy for productivity and temperature to assess whether food webs
were different amongst polar, temperate or tropical regions.
Statistical analysis
To test for differences in food web attributes (number of consumers, number of
producers, FCL and CR) among habitats, zones and locations, a one-factor univariate
permutational analysis of variance (PERMANOVA) was used, based on a Euclidean
distance similarity matrix, which gives equivalent results to a standard ANOVA
(Anderson 2001, McArdle and Anderson 2001) but has no inherent assumptions
regarding normality of data. All permutations were based on raw data and all treatments
were treated as fixed factors. Analyses were performed using Primer-E +
PERMANOVA (v6.1).
Multiple linear regressions were performed to explore the relationships between FCL
and several environmental or food web attributes that: (1) were easily obtainable for the
majority of studies; and (2) may contribute to FCL (Post 2002a). These were: ocean
surface primary productivity (mgC/m2/day); latitude; number of consumers (sampled);
number of producers (sampled); and top predator (maximum) δ15N. Standard linear
regressions were used to determine relationships between number of producers and
CRconsumers or CRproducers. Data were log10 transformed where necessary to satisfy
normality assumptions. All factors were determined as independent variables after
checks for co-linearity.
159
Chapter VI
Results
Comparison among studies (data validation)
We collected stable isotope and food web data from 97 marine food webs that
encompassed several different environments, zones and habitats, with the majority of
studies occurring in temperate latitudes from the northern hemisphere (Figure 1, Table
1, Appendix 1). Benthic marine food webs comprised 76% of studies, those
incorporating both environments 14%, with pelagic studies comprising 10%. Within the
benthic zone, coastal areas (48%), followed by estuarine habitats (39%), were most
frequently studied, with shelf/deep sea areas the least frequent (13%). Amongst habitats,
coastal rocky reef/macroalgae (18%) and coastal sediments (16%) were the most
frequently studied habitats.
There were no statistically significant differences in the number of food sources
sampled amongst environments (benthic, pelagic, both: F2,81=2.8729, p=0.065),
although this is likely an artefact of large variability amongst coastal habitats (Table 1,
Figure 2). Similarly, comparisons across zones just failed to reach significance (coastal,
estuary, shelf/deep sea: F2,62=3.1893, p=0.054) and this was likely a result of the
considerable variation in the number of producers (sampled) within the coastal category.
There were differences amongst benthic habitats (F9,55=2.5484, p=0.019), with the
shelf/deep sea habitat significantly lower than other habitats (Table 1, Figure 2).
There were no significant differences between the number of species sampled amongst
environments (F2,94=1.8132, p=0.177), zones (F2,66=0.86152, p=0.416) or habitats
(F9,58=1.2075, p=0.278), although there was considerable variation amongst means
within each group (Table 1, Figure 2). There were no consistent trends in variability
between number of producers or the number of consumers sampled amongst different
environments, zones or habitats (Figure 2).
Food web metrics
FCL
FCL as estimated by δ15N range, was characterised by normal distributions (Figure 3)
and spanned at least three trophic levels (Table 1). There were no significant difference
in FCL amongst environments (F2,84=0.064, p=0.934), zones (F2,57= 0.912, p=0.392), or
benthic habitats (F9,49=0.967, p=0.474). Multiple regressions revealed the combined
160
Chapter VI
variables of top δ15N and productivity (mgC/m2/day) explained a considerable
proportion of FCL variation, with an R2 value of 0.456. There was a significant, positive
trend of increasing FCL with increasing top δ15N (t=4.119, p=<0.001) and significant,
negative trend of FCL with increasing productivity (productivity: t= -4.107, p<0.001,
Figure 4). There was no trend for latitude (t=1.112, p=0.271) or number of producers or
consumers sampled (producers- t= -0.87, p=0.424; consumers - t= 0.320, p= 0.751).
Figure 1. Map of study locations and productivity estimates of global depth-integrated chlorophyll
biomass (mgC/m2/day) averaged over 10 years from 2003-2010 obtained from SeaWiFS data.
161
Chapter VI
Table 1. Mean values (± SE) of food web metrics derived from 97 isotope food web studies.
Environment
n
# Producers
# Consumers
Top δ N (‰)
15
FCL
CRconsumers
Benthic
74
7.71 ± 0.23
34.73 ± 2.43
13.64 ± 0.38
3.99 ± 0.08
9.4 ± 1.49
Both
13
5.00 ± 0.92
30.69 ± 4.66
15.22 ± 0.64
3.94 ± 0.16
7.29 ± 1.30
Pelagic
10
1.83 ± 0.23
22.3 ± 4.43
12.38 ± 3.24
3.98 ± 0.19
5.28 ± 1.49
Coastal
34
9.00 ± 1.45
34.47 ± 5.83
13.55 ± 0.53
3.89 ± 0.14
8.16 ± 0.60
Estuary
26
7.69 ± 1.07
34.15 ± 5.34
13.84 ± 0.49
3.80 ± 0.15
9.53 ± 0.71
Shelf/deep sea
9
2.44 ± 0.67
44.33 ± 7.12
15.00 ± 0.49
4.15 ± 0.18
8.34 ± 1.12
Coastal sediment
12
4.27 ± 1.36
41.73 ± 9.31
14.83 ± 0.84
3.92 ± 0.17
6.71 ± 0.52
Mangroves
5
7.40 ± 2.40
47.20 ± 7.60
11.62 ± 0.98
3.98 ± 0.33
11.62 ± 0.69
Rocky sediment
10
9.00 ± 2.85
29.40 ± 3.90
14.60 ± 0.79
3.65 ± 0.08
8.88 ± 1.21
Rocky/macroalgae
12
11.58 ± 2.47
35.00 ± 5.89
11.83 ± 0.66
3.88 ± 0.18
8.17 ± 1.25
Saltmarsh
8
7.25 ± 1.24
25.63 ± 5.49
15.32 ± 0.94
4.00 ± 0.20
8.12 ± 0.90
Saltmarsh/mangrove
2
7.00 ± 1.00
18.00 ± 4.00
18.05 ± 1.25
3.49 ± 0.11
6.65 ± 0.55
Sandy beach
2
2.00 ± 0.00
11.00 ± 4.00
16.73 ± 0.16
4.36 ± 0.19
7.95 ± 0.49
Seagrass
7
10.14 ± 2.88
37.71 ± 6.75
11.09 ± 1.19
3.42 ± 0.15
9.68 ± 1.10
Seagrass/mangrove
2
19.50 ± 9.50
35.50 ± 17.50
11.58 ± 0.22
3.90 ± 0.37
11.68 ± 0.67
Shelf/deep sea sediment
9
2.44 ± 0.67
44.33 ± 7.12
15.00 ± 0.49
4.15 ± 0.18
8.34 ± 1.12
Zone (benthic only)
Habitat (benthic only)
162
Chapter VI
CR
As a result of isotopic averaging, CRconsumers should be equal to or less than CRproducers if
the number of food sources sampled were sufficient to characterise the base of the food
web. Twenty-one out of 75 studies (28%) reported CRconsumers greater than CRproducers,
indicating these studies may not have sampled a sufficient range of producers to
characterise available food sources for consumers. Consequently, we removed these
studies before further analysis to avoid bias.
There was a significant increase in CRproducers with increasing number of producers
(sampled) across all studies (F1,74 = 38.171, R2 = 0.343, p<0.001; Figure 5a). There was
also a significant but weaker positive correlation between CRconsumers and number of
producers (F1,53 = 8.576, R2 = 0.142, p= 0.005; Figure 5b). There were no significant
difference for CRconsumers between zones (F2,67 = 1.207, p= 0.285) or habitats (F9,59=
1.456, p= 0.171), yet there was a significant difference between environments (F2,88=
5.189, p= 0.008) with pelagic food webs differing from benthic food webs and food
webs that incorporated both environments (Both). For CRproducers, comparisons between
both environments and habitats just failed to reach significance (environment - F2,72=
2.848, p= 0.058; habitats - F9,48= 1.8767, p= 0.076) but were suggestive of underlying
patterns. There were, however, clear differences between zones (F2,70= 5.4266, p=
0.005) with CRproducers of coastal food webs (14.690 ± 1.19, mean ± SE) being
significantly different compared to estuarine (12.599 ± 1.05) and shelf/deep sea food
webs (9.132 ± 1.86).
163
Chapter VI
Figure 2. Box plots displaying variance amongst studies in the number of consumers and producers
sampled from different environments (top), zones (middle) and habitats (bottom). Boxes are interquartile
ranges, horizontal lines are medians, vertical lines are ranges with outliers (•) set at 95% confidence
intervals.
164
Chapter VI
Figure 3. Frequency distributions (percentage of total) of food chain length using δ15N from different
systems. A=benthic only; B=both (benthic and pelagic); C=pelagic only; D= coastal; E=estuary;
F=Shelf/deep sea. Note: D-F benthic habitats only.
165
Chapter VI
Figure 4. Realtionship between food chain length and primary productivity.
166
Chapter VI
Figure 5. (A) Relationship between number of producers and the carbon range (CR) of producers using
δ13C and, (B) relationship between number of producers and carbon range (CR) of consumers using δ13C.
There was a significant positive correlation of number of producers with CRconsumers and CRproducers.
167
Chapter VI
Discussion
This study set out to determine if data obtained from the SIFWS literature conformed to
broad ecological and community descriptors and if comparative analysis of SIFWS
could be used to identify broad-scale patterns relating to food web theory. We found
clear differences between the numbers of food sources sampled amongst different
environments, zones and habitats. Studies conducted in nearshore environments
generally sampled a greater number of primary producers compared to shelf/deep sea
locations or sedimentary environments. This would be expected in systems that
generally contain greater species diversity in primary producers. Although we found no
statistical patterns relating to differences in the number of species sampled among
different environments, there were considerable indications (differences between
means) suggesting actual differences are likely. Whilst elementary, these results confirm
assumptions regarding the capacity for SIFWS to reflect the ecological and community
factors governing their creation, allowing for cross-system comparisons.
We found that FCL was similar across marine habitats, averaging close to four trophic
levels, yet was variable amongst studies within habitat categories. FCL was not
correlated with latitude, or the number of producers or consumers (sampled), but did
show a significant positive correlation with top δ15N and a significant, albeit weak,
negative correlation with mean ocean primary productivity. CRproducers was positively
correlated with increasing number of producers, indicating that systems characterised by
a large CR are more likely to have a greater number of producers. We also confirm that
CRconsumers is significantly correlated with the number of producers sampled, which
suggests a large CRconsumers is representative of systems with a greater diversity of
carbon sources, as originally hypothesised by Layman et al. (2007). These results imply
that both δ15N and δ13C values, when standardised, are able to provide insight into
ecologically important, large-scale, cross-system patterns of carbon flow and trophic
structure.
Food chain length
Traditional theory argues that FCL is fundamentally governed by either resource
availability (Elton 1927, Hutchinson 1959, Schoener 1989) or dynamic stability (Pimm
and Lawton 1977, Pimm 1982). More recent studies suggest other food web properties
such as ecosystem size (Post et al. 2000), size structure of predator-prey interactions
168
Chapter VI
(Hastings and Conrad 1979) and disturbance (Marks et al. 2000) can also have a
controlling effect on FCL. Whilst no single determinant of FCL has emerged, ecologists
should look to identify which suite of interacting factors most constrain FCL within
each food web type (Post 2002a).
We found that FCL varied with mean primary productivity (mgC/m2/day), although this
was a negative relationship, suggesting systems with higher productivity have shorter
food chains. This was supported when analysing differences in FCL between zones (i.e.
coastal and estuary vs. shelf/deep sea) with shelf/deep sea habitats having, on the whole,
longer FCLs. This result contrasts with traditional (Eltonian) resource availability
theory which implies food chains should be longer in systems with higher productivity
as a result of greater energy being available for transfer between trophic levels
(Hutchinson 1959, Schoener 1989). In his review on factors contributing to FCL, Post
(2002) provides evidence showing that resource availability limits FCL only in systems
with very low levels of productivity. Our results suggest when food becomes a nonlimiting factor, few specialised feeders occur within each trophic level, likely as a result
of low competition. The food web could be considered ‘compressed’, with little trophic
variation within feeding guilds (Figure 6). Under an increasingly limited food supply,
competition within each food level increases, resulting in more specialist feeders in each
feeding guild, overall higher food web trophic diversity, and ‘expanded’ food webs.
Considering no study to date has assessed how FCL varies with primary productivity
across multiple marine systems, we highlight the novel aspect of our findings and
suggest further analysis would shed considerable light on this avenue of theoretical food
web research.
‘Compressed’
Food chain length
‘Expanded’
Increasing carbon availability
Figure 6. Conceptual trophic model for food webs with different amounts of carbon availability. Food
webs become more compressed and have lower within guild diversity with increasing carbon availability.
169
Chapter VI
In the only other study to date that has calculated FCL across different marine
environments using stable isotopes of carbon and nitrogen, Vander Zanden and Fretwell
(2007) found almost identical FCL to those identified in this study (3.97). These authors
also found a similar lack of correlation with latitude, and no difference between
estuarine, coastal and pelagic systems. Our results, and those of Vander Zanden and
Fretwell (2007), suggest that marine food webs may be uniform in the number of
trophic levels, regardless of ecosystem type or habitat. We also found no difference in
FCL between pelagic and benthic food webs. Benthic food webs are thought to have a
lower number of trophic levels (median three) compared to pelagic food webs (median
four) due to the lack of intermediate species (Wiegert and Owen 1971, Schoener 1989).
Yet we found both benthic and pelagic food webs had four trophic levels. Similarly, we
found no differences in FCL amongst habitats with different disturbance regimes, a key
element of the dynamic stability hypothesis (Pimm and Lawton 1977). Shelf and deep
sea systems, considered relatively stable, had similar FCL compared to dynamic
systems such as sandy beaches. Although deep sea systems could be considered more
energy efficient, since they reach similar levels of trophic diversity with considerable
less production and biomass.
Overall, these results highlight the difficulty in determining the factors governing FCL
and suggest much more work is required to identify them. The promising nature of our
results does, however, indicate SIFWS will be of considerable use to future studies
seeking to untangle the factors governing FCL and highlights the need to include studies
from marine habitats when determining factors influencing FCL.
Carbon range
Our purpose was to determine if CR is a useful index for summarising the available
dietary niche diversity of entire food webs from the marine environment. Whilst the
concept of ecological niche is considered fundamental to ecology (Cohen 1978), the
complexity associated with measuring both the biological and physical requirements of
an organism, and the factors that impact on their resources, coupled with confusion
around a standard definition, highlights the considerable difficulty in actually
calculating an organism’s ecological niche (Leibold 1995). Stable isotopes of carbon
and nitrogen have been proposed as one easily obtainable, repeatable measure which
can summarise at least two components of ecological niche: an individual’s dietary and
170
Chapter VI
trophic niche (Bolnick et al. 2002, Bearhop et al. 2004). Two-dimensional plots of δ13C
against δ15N (i.e. diet against trophic position) are comparable to the ecological axes in
Hutchinson’s original niche description (Hutchinson 1957).
However, one of the key limitations when interpreting isotopic variability as dietary
niche diversity is resolving the conflict between differences attributed to variation
assimilated from producers and isotopic variation among individuals due to real
differences in diet. Dietary niche diversity in consumers can only be interpreted as true
dietary differences when the variation in producers has been accounted for (Matthews
and Mazumder 2004). To date, much of the focus of studies on dietary variability has
been at the individual level within single food webs. However, the calculation of
isotopic dietary niche is also relevant to cross-system comparisons, as a measure of
resource availability. Only one study to date has assessed the capacity of stable isotopes
to calculate resource availability for whole food webs (Layman et al. 2007).
Our results suggest that CRconsumers could be a useful index with which to explore
concepts of dietary niche diversity for whole systems across different habitats. We
found that CRconsumers varied between pelagic and benthic food webs, with pelagic food
webs generally having a narrower CRconsumers than benthic food webs. This was
consistent across other food web indices with pelagic food webs mostly having fewer
producers (also represented by narrower CRproducers). However, we express some caution
when interpreting these results. Our analyses of pelagic studies are likely to
underrepresent the total number of different food sources, because many studies only
provided mean values of different microphytes (e.g. recorded as SPOM/Pelagic POM).
This is in contrast to benthic studies, which on the whole tended to provide values for
separate macroalgal species.
Although we found no statistical difference between number of producers or CRconsumers
among different environments or zones, there were obvious differences amongst
habitats. We suggest the lack of any statistical differences may largely be an artefact of
inconsistent sampling amongst studies, particularly amongst coastal environments.
However, it is also possible that some habitats naturally contain more variation in the
number of producers they contain, resulting in weaker power to detect these patterns
statistically.
171
Chapter VI
We note our finding are partially consistent with those of Schoener (1989) who
described pelagic webs as being ‘tall and skinny’ with a larger number of intermediate
species, compared to benthic webs which are ‘short and fat’ with a low number of
intermediate species, but a high number of detritvores. These generalisations break
down, however, when comparing FCL in benthic and pelagic systems. In contrast to the
findings of Schoener (1989) for connectance webs, we found no evidence of tall or short
webs, with consistent FCL across benthic and pelagic systems.
We found CRproducers was significantly positively correlated with increasing number of
producers, indicating that systems characterised by a high CRproducers are more likely to
have a high number of producers. Whilst intuitive, several authors have questioned
whether a narrow CRconsumers is merely representative of a system where many producers
have similar δ13C values, resulting in narrow carbon ranges, rather than a system with
few producers (i.e. reduced niche diversity) (Matthews and Mazumder 2004,
Hoeinghaus and Zeug 2008). Although this may be true when comparing among only
few systems, the results from our study suggest, at least when comparing food webs
across larger scales, there is a consistent positive relationship between CRconsumers and
number of producers. These findings provide confidence that CRconsumers could be used
as one measure of the resource diversity within a system. However, we highlight that
such a simple measure only indicates diversity of basal resources, and does not directly
equate to true resource availability for consumers, a measure that should include
resource parameters such as biomass and palatability. Nonetheless, we have shown that
CRconsumers is a promising index of resource diversity amongst different systems and is
responsive to key food web elements at the system scale.
Comment on SIFWS
Direct comparison of food webs is particularly difficult in marine systems compared to
terrestrial or lake systems, because they often suffer from a lack of clearly described
standard units of measurements. For example, defining ecosystem boundaries in marine
systems is largely subjective, hindering quantitative analysis when determining the
constraining effect of ecosystem size on FCL (Post 2002a, Vander Zanden and Fetzer
2007). Defining food web boundaries is equally difficult: many top predators have wide
foraging areas and the inclusion of top predators such as marine mammals and birds is
haphazard, likely resulting in their under-representation in marine food webs, which
172
Chapter VI
could result in lower FCL (Vander Zanden and Fetzer 2007). SIFWS are prone to other,
isotope-specific, errors such as subjective trophic classification (when there is a lack of
detailed dietary data) and variable fractionation rates among species, resulting in
inherent variation when calculating FCL (Kaehler et al. 2000, Vander Zanden and
Rasmussen 2001, McCutchan et al. 2003). Lastly, benthic species are thought to be
enriched in δ15N compared to pelagic species as a result of lower water turbulence and
consequent recycling of δ15N in static waters (France 1995). The δ15N values among
species occupying or foraging in different habitats may thus be subject to considerable
variation. Direct comparisons among species will therefore require the addition of
dietary information to identify bias associated with benthic enrichment.
Despite almost 20 years elapsing since Cohen et al. (1993) raised concern that the lack
of consistency among food webs studies hindered critical testing of theoretical concepts,
we feel that little has changed. In particular, the failure of some studies to adequately
sample the range of producers within a food web may lead to erroneous conclusions on
diet or trophic position (Matthews and Mazumder 2004). Large discrepancies in the
level of detail reported were also evident in the literature we surveyed, a trait common
across all food web types (Pimm et al. 1991, Polis 1991). Several studies described as
‘food webs’ were not much more than a brief trophic synopsis of several common
species, whilst others went into considerable detail encompassing many species. We
suggest these differences ultimately stem from two different research aims: (1) studies
seeking to determine the diet and trophic position of an individual(s) in the context of a
food web; and (2) studies genuinely seeking to understand carbon flow and trophic
properties within whole communities.
Both types of studies are likely to be advantageous when testing food web theories
because both contain different types of requisite detail. We found those studies which
sampled fewer groups, on the whole, had more detailed dietary information and often
supplemented isotopic data with other dietary methods, whilst many also clearly stated
food web links, thus enabling direct comparison with traditional connectance webs. On
the other hand, those studies which could be considered more comprehensive food web
studies, covered a larger number of trophic groups representing a greater proportion of
species, yet were lacking in specific dietary detail. Studies that are both detailed and
comprehensive are clearly of greatest benefit to food web theory, yet, due to the
173
Chapter VI
considerable effort involved in generating such webs, they are likely to remain
uncommon for the foreseeable future.
The greatest difficulty we had, however, and which we hypothesise as contributing to
some confounding of the patterns presented in our review, was the lack of standardised
and clear reporting of common food web attributes, such as sampling effort, proportion
of species sampled, and clear definitions of food web limitations. We see this as an area
that should be given greater emphasis in future studies. Furthermore, greater reporting
of sampling effort would provide studies such as ours with a greater capacity to
categorise similar studies based on empirical evidence. We therefore reiterate the
sentiments of Cohen et al. (1993) that, if we wish to make advances in testing food web
theory using empirical evidence, ecologists who use stable isotopes for food web studies
must be both more exhaustive and explicit in providing detail when reporting food web
data. Particular attention should be paid to reporting sampling effort (such as yield
effort curves) and standardised metrics (e.g. FCL, CR, top δ15N). Although the reporting
of raw isotopic data is reasonably consistent (e.g. tabulated values of δ13C and δ15N,
δ13C versus δ15N bi-plots), we emphasise that much work still needs to be done to
clearly define the environmental, biological and physical parameters under which the
isotopic data were collected. Without this, we limit our ability to categorise food web
studies based on effort, hindering accurate within- and across-system comparisons.
Failure to include such readily compiled, yet rarely reported, information will continue
to restrict the application of stable isotope analysis to single systems.
174
Chapter VI
Conclusion
We hope this study provokes more thought on the benefits of SIFWS as a resource for
testing food web concepts. The body of literature that contains isotopic data on trophic
and feeding links is now too large to be ignored by food web theorists, yet, if SIFWS
are to play a role in advancing food web theory, we must identify the situation in which
they will be most useful. In this respect, we still have much to determine. For example,
it is unclear if sampling methodologies in SIFWS reflect those of traditional
connectance webs such that community patterns (e.g. constant proportion of species
within trophic groups, an even ratio of predator to prey species) align with those
reported for traditional webs (Pimm 1980, Schoener 1989). Furthermore, it is unclear if
SIFWS share the same type of bias as traditional webs (e.g. aggregation of species).
Despite the unknowns, we have shown that comparative analysis of SIFWS has
considerable merit and the evidence from this study suggests that food web metrics
derived from SIFWS are responsive to key food web properties. Greater emphasis must
be placed on standardising metrics and clear reporting of food web attributes and
sampling effort if patterns are to be more clearly delineated in the future.
175
Chapter VI
Acknowledgements
I thank Glenn Johnstone who provided valuable advice on earlier versions of this
manuscript. Ben Raymond provided statistical advice and supplied primary productivity
data and Figure 1, in conjunction with the Australian Antarctic Data Centre. This
research was funded by a PhD research scholarship in part by University of New
England and Southern Cross University, and supported financially and logistically by
the Australian Antarctic Division (AAS projects 2948 and 2201).
176
Chapter VI
References
Akin S, Winemiller KO (2008) Body size and trophic position in a temperate estuarine
food web. Acta Oecolo 33:144-153
Anderson MJ (2001) A new method for non-parametric multivariate analysis of
variance. Austral Ecol 26:32-46
Bearhop S, Adams CE, Waldron S, Fuller RA, Macleod H (2004) Determining trophic
niche width: a novel approach using stable isotope analysis. J Anim Ecol 73:1007-1012
Bolnick DI, Yang LH, Fordyce JA, Davis JM, Svanback R (2002) Measuring
individual-level resource specialization. Ecology 83:2936-2941
Brose U, Jonsson T, Berlow EL, Warren P, Banasek-Richter C, Bersier L-F, Blanchard
JL, Brey T, Carpenter SR, Blandenier M-FC, Cushing L, Dawah HA, Dell T, Edwards
F, Harper-Smith S, Jacob U, Ledger ME, Martinez ND, Memmott J, Mintenbeck K,
Pinnegar JK, Rall BC, Rayner TS, Reuman DC, Ruess L, Ulrich W, Williams RJ,
Woodward G, Cohen JE (2006) Consumer-resource body-size relationships in natural
food webs. Ecology 87:2411-2417
Cabana G, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen
isotopes. Proc Nat Acad Sci USA 93:10844-10847
Carpenter SR, Kitchell JF, Hodgson JR (1985) Cascading trophic interactions and lake
productivity. Biosci 35:634-639
Chikaraishi Y, Ogawa N, Kashiyama Y, Takano Y, Suga H, Tomitani A, Miyashita H,
Kitazato H, Ohkouchi N (2009) Determination of aquatic food-web structure based on
compound-specific nitrogen isotopic composition of amino acids. Limnol Oceanogr
Methods 7:740-750
Closs G. (1991) Multiple definitions of food web statistics: An unnecessary problem for
food web research. Aust J Ecolo 16:413-415
Cohen JE (1978) Food webs and niche space, Vol. Princeton University Press,
Princeton, New Jersey
Cohen JE, Beaver RA, Cousins SH, Deangelis DL, Goldwasser L, Heong KL, Holt RD,
Kohn AJ, Lawton JH, Martinez N, O'malley R, Page LM, Patten BC, Pimm SL, Polis
177
Chapter VI
GA, RejmãNek M, Schoener TW, Schoenly K, Sprules WG, Teal JM, Ulanowicz RE,
Warren PH, Wilbur HM, Yodzis P (1993) Improving food webs. Ecology 74:252-258
Deniro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in
animals. Geochimica et Cosmochimica Acta 45:341-351
Dunne J, Brose U, Williams RJ, Martinez ND (2005) Modelling food web dynamics:
complexity-stability implications. In: Belgrano A, Scharler UM, Dunne J, Ulanowicz
RE (eds) Aquatic food webs: an ecosystem approach. Oxford university press, Oxford
Elton C (1927) Animal ecology, 2nd ed. Sidgwick and Jackson, London
Fantle MS, Dittel AI, Schwalm SM, Epifanio CE, Fogel ML (1999) A food web
analysis of the juvenile blue crab, Callinectes sapidus, using stable isotopes in whole
animals and individual amino acids. Oecologia 120:416-426
France RL (1995) Differentiation between littoral and pelagic food webs in lakes using
stable carbon isotopes. Limnol Oceanogr 40:1310-1313
Fretwell SD (1987) Food chain dynamics: the central theory of ecology? Oikos 50:291301
Gillies CL Stark JS Smith SDA (2012) Small-scale spatial variation of δ13C and δ15N
isotopes in Antarctic carbon sources and consumers. Polar Biol :1-15 In Press
Hastings HM, Conrad M (1979) Length and evolutionary stability of food chains.
Nature 282:838-839
Hoeinghaus DJ, Zeug SC (2008) Can stable isotope ratios provide for community-wide
measures of trophic structure? Comment. Ecology 89:2353-2357
Hutchinson GE (1957) Concluding remarks. Cold Spring Harbour Symposia. Quant
Biol 22:415-427
Hutchinson GE (1959) Homage to Santa Rosalia or why are there so many kinds of
animals? Amer Nat 93:145-159
Jennings S, Pinnegar JK, Polunin NVC, Boon TW (2001) Weak cross-species
relationships between body size and trophic level belie powerful size-based trophic
structuring in fish Communities. J Animal Ecol 70:934-944
178
Chapter VI
Kaehler S, Pakhomov EA, McQuaid CD (2000) Trophic structure of the marine food
web at the Prince Edward Island (Southern Ocean) determined by δ13C and δ15N
analysis. Mar Ecol Prog Ser 208:13-20.
Layman C, Arrington D, Montaña C, Post D (2007) Can stable isotope ratios provide
for community-wide measures of trophic structure. Ecology 88:42-48
Leibold MA (19950 The niche concept revisited: mechanistic models and community
context. Ecology 76:1371-1382
Marks JC, Power ME, Parker MS (2000) Flood disturbance, algal productivity, and
interannual variation in food chain length. Oikos 90:20-27
Matthews B, Mazumder A (2004) A critical evaluation of intrapopulation variation of
δ13C and isotopic evidence of individual specialization. Oecologia 140:361-371
May RM (1973) Stability and complexity in model ecosystems. Princeton University
Press, Princeton
McArdle BH Anderson MJ 2001 Fitting multivariate models to community data: a
comment on distance based redundancy analysis. Ecology 82:290-297
McClelland JW, Montoya JP (2002) Trophic relationships and the nitrogen isotopic
composition of amino acids in plankton. Ecology 83:2173-2180.
McCutchan JH, Lewis WM, Kendall C, McGrath CC (2003) Variation in trophic shift
for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102:378-390
Moore JW, Semmens BX (2008) Incorporating uncertainty and prior information into
stable isotope mixing models. Ecol Lett 11:470-480
Newsome S, Martinez Del Rio C, Bearhop S, Phillips D (2007) A niche for isotopic
ecology. Front ecol env 5:429-436
Oksanen L, Fretwell SD, Arruda J, Niemela P (1981) Exploitation ecosystems in
gradients of primary productivity. Amer Nat 118:240-261
Paine RT (1988) Road maps of interactions or grist for theoretical development?
Ecology 69: 1648-1654
Pasquaud S, Lobry J, Elie P (2007) Facing the necessity of describing estuarine
ecosystems: a review of food web ecology study techniques. Hydrobiologia 588:159172
179
Chapter VI
Peterson, BJ (1999) Stable isotopes as tracers of organic matter input and transfer in
benthic food webs: a review. Acta Oecolo 20:479-487
Phillips D, Newsome S, Gregg J (2005) Combining sources in stable isotope mixing
models: alternative methods. Oecologia 144:520-527
Pimm SL (1980) Properties of food webs. Ecology 61:219-225
Pimm SL (1982) Food webs. Chapman and Hall, New York
Pimm SL, Lawton JH (1977) Number of trophic levels in ecological communities.
Nature 268:329-331
Pimm SL, Lawton JH, Cohen JE (1991) Food web patterns and their consequences.
Nature 350:669-674
Polis GA (1991) Complex trophic interactions in deserts: an empirical critique of foodweb theory. Amer Nat 138:123-155
Post DM (2002a) The long and short of food-chain length. Trends Ecol Evol 17:269277
Post DM (2002b) Using stable isotopes to estimate trophic position: models, methods,
and assumptions. Ecology 83:703-718
Post DM, Pace ML, Hairston JNG (2000) Ecosystem size determines food-chain length
in lakes. Nature 405:1047
Raymond B, Marshall M, Nevitt G, Gillies CL, Van Den Hoff J, Stark JS, Losekoot M,
Woehler EJ, Constable AJ (2011) A Southern Ocean dietary database. Ecology
92:1188-1188
Schoener TW (1989) Food webs from the small to the large: the Robert H. MacArthur
Award lecture. Ecology 70:1559-1589
Vander Zanden MJ, Fetzer WW (2007) Global patterns of aquatic food chain length.
Oikos 116:1378-1388
Vander Zanden MJ, Rasmussen JB (1999) Primary consumer δ13C and δ15N and the
trophic position of aquatic consumers. Ecology 80:1395-1404
180
Chapter VI
Vander Zanden MJ, Rasmussen JB (2001) Variation in δ15N and δ13C trophic
fractionation: implications for aquatic food web studies. Limnol Oceanogr 46:20612066
Wada E, Mizutani H, Minagawa M (1991) The use of stable isotopes for food web
analysis. Crit Rev Food Sci Nutr 30:361-371
Wiegert RG, Owen DF (1971) Trophic structure, available resources and population
density in terrestrial vs. aquatic ecosystems. J Theor Biol 30:69-81
181
Chapter VI
Appendix
Appendix 1 (over page). Summary of 97 marine food webs reviewed in this study.
182
Chapter VI
Reference #
Location
# producers
# consumers
CRproducers
Top predator name
Top predator
δ15N (‰)
FCL
CRconsumers
1
Windmill Islands
11
32
26.40
Notothenia coriiceps
13.34
4.20
13.95
2
Vestfold Hills
9
82
21.34
Trematomus bernacchii
13.98
4.10
13.61
3
Palmer station
22
34
23.80
Notothenia coriiceps
12.00
4.47
13.00
4
Bouvet Island
7
33
23.85
Parborlasia corrugatus
10.30
4.26
14.21
5
Prince Edward Islands
3
10
14.17
Lepidonotothen larseni
8.00
4.94
6.00
6
Prince Edward Islands
3
12
14.17
Aglaophamus ornatus
8.00
3.76
4.00
7
Prince Edward Islands
3
16
14.17
Cucumaria Kerguelensis
11.50
5.09
7.00
8
West Antarctic Peninsula
3
27
14.97
Ctenocidaris perrieri
15.60
4.21
8.51
9
Weddell Sea
-
29
-
Pagothenia borchgrevinki
7.10
3.41
9.30
10
Southern Ocean
2
12
5.50
Pseudochirella mawsoni
9.80
4.09
5.70
11
Chukchi Sea
1
48
-
Crossaster papposus
19.20
4.76
6.30
12
Chukchi Sea
1
22
-
Crossaster papposus
18.30
4.35
5.20
13
Chukchi Sea
1
26
-
Unidentified sculpin
16.60
3.91
8.70
14
Canada Basin
2
28
1.10
Hymenodora glacialis
16.60
4.11
7.20
15
Canada Basin
2
35
1.10
Glycinde wireni
17.70
5.55
10.60
16
North water Polynya
3
45
18.40
Raja radiata
18.80
5.21
12.30
17
Barrow Strait-Lancaster
Sound
6
27
2.20
Gymnellus virides
16.20
4.06
5.80
18
Barents Sea
2
50
5.68
Poraniomorpha tumida
17.40
3.97
11.60
19
Porcupine Abyssal Plain
1
90
-
Hexactinellidae
19.93
4.23
8.26
20
Barra del Chuy
2
15
15.13
Hemipodus olivieri
16.56
4.55
8.43
21
Arachania
2
7
17.30
Hemipodus olivieri
16.89
4.17
7.46
22
Bay of Banyuls-sur-Mer
7
121
20.10
Diplodus sargus
12.60
4.09
10.60
23
Gulf of St Lawrence
3
63
9.10
Cottidae
17.00
4.35
5.90
24
Río de la Plata
8
42
14.50
Galeorhinus galeus
19.50
4.94
9.00
25
Gulf of Lions
5
21
5.36
Solea impar
10.93
3.42
5.27
183
Chapter VI
26
Tijuana Estuary
12
24
11.90
Atherinops affinis
16.80
3.71
7.90
27
St Lawrence Estuary
3
59
9.10
Urophycis tenuis
16.70
4.37
7.00
28
Georges Bank
1
38
-
Lophius americanus
14.00
4.15
6.30
29
Patos-Mirim Lagoon
6
17
16.86
Netuma barba
15.51
4.31
11.57
30
Deluge Inlet
4
67
11.00
Scomberoides lysan
12.10
3.76
9.10
31
Andhra Pradesh
2
38
4.00
Metapenaeus monoceras
14.50
5.06
11.50
32
Bering Sea
1
29
-
-
-
-
5.30
33
King George Island
5
25
8.90
-
-
-
14.60
34
Sapelo Island
4
20
17.00
Chaetodipterus faber
12.00
3.24
4.85
35
Barents Sea
2
17
4.60
Sagitta elegans
12.18
3.29
2.24
36
Juan de Fuca Ridge (hydrothermal vent)
35
Nemertea sp.1
9.00
4.31
19.50
37
Bering Sea
-
32
-
Gadus macrocephalus
16.20
3.47
8.20
38
Chukchi Sea
-
50
-
Buccinum sp.
15.20
3.21
5.50
39
Eastern Beaufort Sea
-
35
-
Anonyx sarsi/nugax
15.40
3.76
6.20
40
False Creek harbour
4
14
5.60
Leptocottus armatus
14.80
3.85
8.00
41
Bay of Calvi
7
35
25.20
Apogon imberbis
9.77
4.01
4.25
42
North Sea
-
43
-
Trachurus trachurus
17.90
3.94
-
43
Celtic Sea
-
48
-
Merlangius merlangus
17.20
4.06
-
44
Norwegian Sea
14
26
17.73
Pollachius virens
13.22
4.32
14.81
45
Gamo Lagoon
3
13
5.50
Grandidierella japonica
16.70
4.15
4.20
46
Plum Island Sound
9
16
18.70
Gasterosteus wheatlandi
12.50
3.38
7.80
47
Plum Island Sound
4
10
9.80
Osmeus mordax
14.00
3.82
7.00
48
Plum Island Sound
5
12
9.10
Lepomis macrochirus
11.30
3.03
7.90
49
Plum Island Sound
2
19
1.00
Brevoortia tyrannus
14.40
3.94
7.10
50
Horn Island
5
67
9.60
Lolliguncula brevis
15.70
3.68
7.50
51
Graveline Bay
4
56
10.10
Anchoa nasuta
12.40
3.50
10.00
52
Chesapeake Bay
11
11
11.76
-
-
-
7.60
184
Chapter VI
53
Gazi Bay
10
18
13.48
Tylosurus crocodilus
11.36
4.27
12.35
54
Seto Inland Sea
3
33
7.00
Trichiurus japonicus
17.50
3.74
4.20
55
Plesant Bay
2
33
2.60
Nephtys sp
11.90
2.99
6.30
56
Pearl River Estuary
7
21
8.81
Johnius fasciatus
13.30
3.65
6.05
57
Buloh River Estuary
8
48
12.30
-
-
-
13.20
58
Buloh River Estuary
8
20
12.30
-
-
-
11.20
59
Buloh River Estuary
1
77
n/a
-
-
-
6.00
60
St. Croix
23
27
15.90
-
-
-
7.40
61
Cayos Miskitos
14
19
20.00
-
-
-
12.00
62
NW Mediterranean
4
36
13.55
Molpadia musculusy
12.11
3.71
6.46
63
Green Canyon
5
28
13.10
Paraonidae
6.60
4.97
30.40
64
Alaminos Canyon
3
10
8.80
Sipunculida
7.90
5.76
16.00
65
Atwater Valley
4
19
11.20
Anthozoa
12.20
6.62
21.60
66
NE Pacific
1
25
-
Galeorhinus galeus
15.40
3.74
5.70
67
Atlantic Ocean
1
13
-
Nassarius cabrierensis ovoideus
12.95
3.13
3.13
68
Sulu Sea
-
20
-
Bathygadus sp. 1
13.19
3.39
3.17
69
Celebes Sea
-
13
-
Ectreposebastes imus
11.94
3.31
2.14
70
Yellow Sea
1
12
-
Conger myriaster
13.90
4.50
4.10
71
Caribbean Sea
15
57
19.77
Atherinomorus stipes
9.71
3.79
13.92
72
Salses-Leucate Lagoon
7
53
10.30
Atherina boyeri
13.30
3.71
21.20
73
Åland Islands
8
12
12.80
Åland Islands
9.77
3.65
4.80
74
Sørfjord
12
47
18.90
Gadus morhua
16.10
4.06
6.60
75
South China Sea
-
63
-
Thamnaconus hypargyreus
14.70
3.31
5.90
76
East Sea
32
64
23.40
Sebastes oblongus
12.03
3.09
6.30
77
Flamengo Sound
16
31
18.20
Eucinostomus gula
13.30
3.10
4.70
78
Bohai Bay
1
14
-
Lateolabras japonicus
13.28
4.10
7.83
79
Eastern Australia
-
57
-
Isurus oxyrinchus
15.70
4.53
7.20
185
Chapter VI
80
Sine Saloum estuary
5
25
8.50
Elops lacerta
13.00
4.57
9.30
81
Laguna de Rocha
10
22
12.20
Cyphocharax voga
14.30
4.14
9.80
82
Northern Brittany
29
45
22.10
-
-
-
12.70
83
Northern Brittany
2
24
1.10
-
-
-
5.70
84
Gulf of California
6
22
17.70
Urobatis halleri
16.80
3.38
6.10
85
Gulf of California
8
14
14.30
Ocypode occidentalis
19.30
3.60
7.20
86
Gazi Bay
29
53
17.40
Taeniura lymna
11.80
3.53
11.00
87
Tabasco
7
24
3.87
Atractosteus tropicus
10.16
3.49
11.85
88
Tabasco
9
38
7.39
Batrachoides goldmani
10.26
2.92
10.22
89
Comau Fjord
27
19
22.00
Merluccius gayi
19.40
4.59
8.00
90
Loxahatchee
6
28
19.60
Lutjanus jocu
13.30
3.88
13.40
91
Shark Bay
16
59
16.00
Sillago schomburgkii
9.70
3.58
12.47
92
Bering Sea
3
15
3.80
Priapulus caudatus
16.50
3.85
5.20
93
Bay of Brest
2
38
7.30
Marphysa sanguinea
15.70
3.79
5.90
94
Ile Verte
11
21
5.05
Actinia equina
11.46
3.19
3.97
95
Penmarc'h
11
18
6.40
Bunodactis verrucosa
12.90
3.10
4.90
96
Revellata Bay
14
42
23.10
Syllidae
8.10
3.44
6.30
97
Revellata Bay
3
32
8.60
Isopods
8.70
3.28
11.00
186
Chapter VI
#
Reference
1-2
Gillies et al. 2012 (chapters IV and IV)
3
Dunton KH (2001) δ15N and δ13C measurements of Antarctic Peninsula fauna: trophic relationships and assimilation of benthic seaweeds. Amer Zoolo 41:99-112.
4
Jacob U, Brey T, Fetzer I, Kaehler S, Mintenbeck K, Dunton K, Beyer K, Struck U, Pakhomov EA, Arntz WE (2006) Towards the trophic structure of the Bouvet Island marine ecosystem. Polar
Biol 29:106-113
5-7
Kaehler S, Pakhomov EA, McQuaid CD (2000) Trophic structure of the marine food web at the Prince Edward Island (Southern Ocean) determined by δ13C and δ15N analysis. Mar Ecol Prog
Ser 208:13-20
8
Mincks SL, Smith CR, Jeffreys RM, Sumida PYG (2008) Trophic structure on the West Antarctic Peninsula shelf: detritivory and benthic inertia revealed by δ13C and δ15N analysis. Deep sea
Res Pt II 55:2502-2514.
9
Rau GH, Hopkins TL, Torres JJ (1991) 15N/14N and 13C/12C in Weddell Sea invertebrates: implications for feeding diversity. Mar Ecol Prog Ser 77:1-6
10
Wada E, Terazaki M, Kabaya Y, Nemoto T (1987) 15N and 13C abundances in the Antarctic Ocean with emphasis on the biogeochemical structure of the food web. Deep Sea Res Part I
34:829-841
11-13
Feder H, Iken K, Blanchard A, Jewett S, Schonberg S (2011) Benthic food web structure in the south eastern Chukchi Sea: an assessment using δ13C and δ15N analyses. Polar Biol 34:521532.
14-15
Iken K, Bluhm BA, Gradinger R (2005) Food web structure in the high Arctic Canada Basin: evidence from δ13C and δ15N analysis. Polar Biol 28:238-249
16
Hobson KA, Fisk A, Karnovsky N, Holst M, Gagnon J-M, Fortier M (2002) A stable isotope (δ13C and δ15N) model for the North Water food web: implications for evaluating trophodynamics and
the flow of energy and contaminants. Deep sea Res Pt II 49:5131-5150
17
Hobson KA, Welch HE (1992) Determination of trophic relationships within a high Arctic marine food web using δ13C and δ15N analysis. Mar Ecol Prog Ser 84:9-18
18
Tamelander T, Renaud PE, Hop H, Carroll ML, Ambrose Jr WG, Hobson KA (2006) Trophic relationships and pelagic–benthic coupling during summer in the Barents Sea Marginal Ice Zone,
revealed by stable carbon and nitrogen isotope measurements. Mar Ecol Prog Ser 310:33-46
19
Iken K, Brey T, Wand U, Voigt J, Junghans P (2001) Food web structure of the benthic community at the Porcupine Abyssal Plain (NE Atlantic): A stable isotope analysis. Prog Oceanogr
50:383-405
187
Chapter VI
20-21
Bergamino L, Lercari D, Defeo O (2011) Food web structure of sandy beaches: Temporal and spatial variation using stable isotope analysis. Estuar Coast Shelf Sci 91:536-543
22
Carlier A, Riera P, Amouroux J-M, Bodiou J-Y, Gremare A (2007) Benthic trophic network in the Bay of Banyuls-sur-Mer (northwest Mediterranean, France): An assessment based on stable
carbon and nitrogen isotopes analysis. Estuar Coast Shelf Sci 72:1-15
23
Lesage V, Hammill MO, Kovacs KM (2001) Marine mammals and the community structure of the estuary and Gulf of St Lawrence, Canada: evidence from stable isotope analysis. Mar Ecol
Prog Ser 210:203-221
24
Botto F, Gaitán E, Mianzan H, Acha M, Giberto D, Schiariti A, Iribarne O (2011) Origin of resources and trophic pathways in a large SW Atlantic estuary: An evaluation using stable isotopes.
Estuar Coast Shelf Sci 92:70-77
25
Darnaude AM, Salen-Picard C, Polunin NVC, Harmelin-Vivien ML (2004) Trophodynamic linkage between river runoff and coastal fishery yield elucidated by stable isotope data in the Gulf of
Lions (NW Mediterranean). Oecologia 138:325-332
26
Kwak TJ, Zedler JB (1997) Food web analysis of southern California coastal wetlands using multiple stable isotopes. Oecologia 110:262-277
27
Lesage V, Hammill MO, Kovacs KM (2001) Marine mammals and the community structure of the estuary and Gulf of St Lawrence, Canada: evidence from stable isotope analysis. Mar Ecol
Prog Ser 210:203-221
28
Fry B (1988) Food web structure on Georges Bank from stable C, N, and S isotopic compositions. Limnol Oceanogr 33:1182-1190
29
Garcia AM, Hoeinghaus DJ, Vieira JP, Winemiller KO (2007) Isotopic variation of fishes in freshwater and estuarine zones of a large subtropical coastal lagoon. Estuar Coast Shelf Sci 73:399408
30
Abrantes K, Sheaves M (2009) Food web structure in a near-pristine mangrove area of the Australian Wet Tropics. Estuar Coast Shelf Sci 82:597-607
31
Bouillon S, Raman AV, Dauby P, Dehairs F (2002) Carbon and nitrogen stable isotope ratios of subtidal benthic invertebrates in an estuarine mangrove ecosystem (Andhra Pradesh, India).
Estuar Coast Shelf Sci 54:901-913
32
McConnaughey T, McRoy CP (1979) Food-web structure and the fractionation of carbon isotopes in the Bering Sea. Mar Biol 53 (3):257-262
33
Corbisier T, Petti MV, Skowronski RP, Brito TS (2004) Trophic relationships in the nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C stable-isotope analysis. Polar Biol
27:75-82
34
Peterson BJ, Howarth RW (1987) Sulfur, carbon, and nitrogen isotopes used to trace organic matter flow in the salt-marsh estuaries of Sapelo Island, Georgia. Limnol Oceanogr 32:11951213
188
Chapter VI
35
Søreide JE, Hop H, Carroll ML, Falk-Petersen S, Hegseth EN (2006) Seasonal food web structures and sympagic-pelagic coupling in the European Arctic revealed by stable isotopes and a
two-source food web model. Prog Oceanogr 71 (1):59-87
36
Bergquist DC, Eckner JT, Urcuyo IS, Cordes EE, Hourdez S, Macko SA, Fisher CR (2007) Using stable isotopes and quantitative community characteristics to determine a local hydrothermal
vent food web. Mar Ecol Prog Ser 330:49-65
37-39
Dunton KH, Saupe SM, Golikov AN, Schell DM, Schonberg SV (1989) Trophic relationships and isotopic gradients among arctic and subarctic marine fauna. Mar Ecol Prog Ser 56:89-97
40
Mackintosh CE, Maldonado J, Hongwu J, Hoover N, Chong A, Ikonomou MG, Gobas FAPC (2004) Distribution of phthalate esters in a marine aquatic food web: comparison to
polychlorinated biphenyls. Environm Sci Technol 38:2011-2020
41
Pinnegar JK, Polunin NVC (2000) Contributions of stable-isotope data to elucidating food webs of Mediterranean rocky littoral fishes. Oecologia 122 (3):399-409
42-43
Jennings S, Pinnegar JK, Polunin NVC, Boon TW (2001) Weak cross-species relationships between body size and trophic level belie powerful size-based trophic structuring in fish
communities. J Anim Ecolo 70:934-944
44
Fredriksen S (2003) Food web studies in a Norwegian kelp forest based on stable isotope (δ13C and δ15N) analysis. Mar Ecol Prog Ser 260:71-81
45
Wada E, Kabaya Y, Kurihara Y (1993) Stable isotopic structure of aquatic ecosystems. J Biosci (Bangalore) 18:483-499
46-49
Deegan L, Garritt R (1997) Evidence for spatial variability in estuarine food webs. Mar Ecol Prog Ser 147:31-47
50
Moncreiff CA, Sullivan MJ (2001) Trophic importance of epiphytic algae in subtropical seagrass beds: evidence from multiple stable isotope analyses. Mar Ecol Prog Ser 215:93-106
51
Sullivan MJ, Moncreiff CA (1990) Edaphic algae are an important component of salt marsh food-webs: evidence from multiple stable isotope analyses Mar Ecol Prog Ser 62:149-159
52
Stribling JM, Cornwell JC (1997) Identification of important primary producers in a Chesapeake Bay tidal creek system using stable isotopes of carbon and sulphur. Estuaries 20:77-85
53
Marguillier S, van der Velde G, Dehairs F, Hemminga M, Rajagopal S (1997) Trophic relationships in an interlinked mangrove-seagrass ecosystem as traced by δ13C and δ15N. Mar Ecol Prog
Ser 151:115-121
54
Takai N, Mishima Y, Yorozu A, Hoshika A (2002) Carbon sources for demersal fish in the Western Seto inland Sea, Japan, examined by δ13C and δ15N analyses. Limnol Oceanogr 47:730741
55
Carmichael RH, Rutecki D, Annett B, Gaines E, Valiela I (2004) Position of horseshoe crabs in estuarine food webs: N and C stable isotopic study of foraging ranges and diet composition. J
Exp Biol Ecol 299:231-253
189
Chapter VI
56
Lee S (2000) Carbon dynamics of Deep Bay, eastern Pearl River estuary, China. II: Trophic relationship based on carbon- and nitrogen-stable isotopes. Mar Ecol Prog Ser 205:1-10
57-59
Rodelli MR, Gearing JN, Gearing PJ, Marshall N, Sasekumar A (1984) Stable isotope ratio as a tracer of mangrove carbon in Malaysian ecosystems. Oecologia 61:326-333
60-61
Fry B, Lutes R, Northam M, Parker PL, Ogden J (1982) A 13C/12C comparison of food webs in Caribbean seagrass meadows and coral reefs. Aquatic Bot 14:389-398
62
Fanelli E, Papiol V, Cartes JE, Rumolo P, Brunet C, Sprovieri M (2011) Food web structure of the epibenthic and infaunal invertebrates on the Catalan slope (NW Mediterranean): Evidence
from δ13C and δ15N analysis. Deep Sea Res Pt I 58:98-109
63-65
Demopoulos AWJ, Gualtieri D, Kovacs K (2010) Food-web structure of seep sediment macrobenthos from the Gulf of Mexico. Deep sea Res Pt II 57:1972-1981
66
Miller TW, Brodeur RD, Rau G, Omori K (2010) Prey dominance shapes trophic structure of the northern California Current pelagic food web: evidence from stable isotopes and diet analysis.
Mar Ecol Prog Ser 420:15-26
67
Sampaio L, Rodrigues AM, Quintino V (2010) Carbon and nitrogen stable isotopes in coastal benthic populations under multiple organic enrichment sources. Mar Pol Bull 60:1790-1802
68-69
Asante KA, Agusa T, Kubota R, Mochizuki H, Ramu K, Nishida S, Ohta S, Yeh H-m, Subramanian A, Tanabe S (2010) Trace elements and stable isotope ratios (δ13C and δ15N) in fish from
deep-waters of the Sulu Sea and the Celebes Sea. Mar Pol Bull 60:1560-1570
70
Wan R, Wu Y, Huang L, Zhang J, Gao L, Wang N (2010) Fatty acids and stable isotopes of a marine ecosystem: study on the Japanese anchovy (Engraulis japonicus) food web in the Yellow
Sea. Deep sea Res Pt II 57:1047-1057.
71
Olsen YS, Fox SE, Kinney EL, Teichberg M, Valiela I (2010) Differences in urbanization and degree of marine influence are reflected in δ13C and δ15N of producers and consumers in seagrass
habitats of Puerto Rico. Mar Environ Res 69:198-206
72
Carlier A, Riera P, Amouroux JM, Bodiou JY, Desmalades M, Gremare A (2009) Spatial heterogeneity in the food web of a heavily modified Mediterranean coastal lagoon: Stable isotope
evidence. Aquatic Biol 5:167-179
73
Nordstrom M, Aarnio K, Bonsdorff E (2009) Temporal variability of a benthic food web: patterns and processes in a low-diversity system. Mar Ecol Prog Ser 378:13-26
74
Nilsen M, Pedersen T, Nilssen EM, Fredriksen S (2008) Trophic studies in a high-latitude fjord ecosystem - a comparison of stable isotope analyses (δ13C and δ15N) and trophic-level
estimates from a mass-balance model. Can J Fish Aquat Sci 65:2791-2806
75
Asante KA, Agusa T, Mochizuki H, Ramu K, Inoue S, Kubodera T, Takahashi S, Subramanian A, Tanabe S (2008) Trace elements and stable isotopes (δ13C and δ15N) in shallow and deepwater organisms from the East China Sea. Environ Pol 156:862-873
190
Chapter VI
76
Kang C-K, Choy E, Son Y, Lee J-Y, Kim J, Kim Y, Lee K-S (2008) Food web structure of a restored macroalgal bed in the eastern Korean peninsula determined by C and N stable isotope
analyses. Mar Biol 153:1181-1198
77
Corbisier TN, Soares LSH, Petti MAV, Muto EY, Silva MHC, McClelland J, Valiela I (2006) Use of isotopic signatures to assess the food web in a tropical shallow marine ecosystem of south
eastern Brazil. Aquat Ecol 40:381-390
78
Wan I, Hu J, An L, An W, Yang M, Mitsuaki I, Tatsuya H, Tao S (2005) Determination of trophic relationships within a Bohai Bay food web using stable δ15N and δ13C analysis. C Sci Bull
50:1021-1025
79
Revill A, Young J, Lansdell M (2009) Stable isotopic evidence for trophic groupings and bio-regionalization of predators and their prey in oceanic waters off eastern Australia. Mar Biol
156:1241-1253
80
Faye D, Tito de Morais L, Raffray J, Sadio O, Thiaw OT, Le Loc'h F (2011) Structure and seasonal variability of fish food webs in an estuarine tropical marine protected area (Senegal):
Evidence from stable isotope analysis. Estuar Coast Shelf Sci 92:607-617.
81
Rodríguez-Graña L, Calliari D, Conde D, Sellanes J, Urrutia R (2008) Food web of a SW Atlantic shallow coastal lagoon: spatial environmental variability does not impose substantial changes
in the trophic structure. Mar Ecol Prog Ser 362:69-83
82-83
Dauby P, Khomsi A, Bouquegneau J-M (1998) Trophic relationships within intertidal communities of the Brittany Coasts: a stable carbon isotope analysis. J Coast Res 14:1202-1212
84-85
Serrano-Grijalva L, Sánchez-Carrillo S, Angeler DG, Sánchez-Andrés R, Álvarez-Cobelas M (2011) Effects of shrimp-farm effluents on the food web structure in subtropical coastal lagoons. J
Exp Mar Biol Ecol 402:65-74
86
Nyunja J, Ntiba M, Onyari J, Mavuti K, Soetaert K, Bouillon S (2009) Carbon sources supporting a diverse fish community in a tropical coastal ecosystem (Gazi Bay, Kenya). Estuar Coast
Shelf Sci 83:333-341
87-88
Mendoza-Carranza M, Hoeinghaus DJ, Garcia A, Romero-Rodriguez A (2010) Aquatic food webs in mangrove and seagrass habitats of Centla Wetland, a Biosphere Reserve in south eastern
Mexico. Neotrop Ichthyolo 8:171-178
89
Mayr C, Försterra G, Häussermann V, Wunderlich A, Grau J, Zieringer M, Altenbach A (2011) Stable isotope variability in a Chilean fjord food web: implications for N- and C-cycles. Mar Ecol
Prog Ser 428:89-104
90
Yeager L, Layman C (2011) Energy flow to two abundant consumers in a subtropical oyster reef food web. Aquat Ecol 45:267-277
91
Heithaus E, Heithaus P, Heithaus M, Burkholder D, Layman C (2011) Trophic dynamics in a relatively pristine subtropical fringing mangrove community. Mar Ecol Prog Ser 428:49-61.
191
Chapter VI
92
Lovvorn JR, Cooper LW, Brooks ML, De Ruyck CC, Bump JK, Grebmeier JM (2005) Organic matter pathways to zooplankton and benthos under pack ice in late winter and open water in late
summer in the north-central Bering Sea. Mar Ecol Prog Ser 291:135-150
93
Grall J, Le Loc'h F, Guyonnet B, Riera P (2006) Community structure and food web based on stable isotopes (δ15N and δ13C) analysis of a North Eastern Atlantic maerl bed. J Exp Mar Biol
Ecol 338:1-15
94-95
Golléty C, Riera P, Davoult D (2010) Complexity of the food web structure of the Ascophyllum nodosum zone evidenced by a δ13C and δ15N study. J Sea Res 64:304-312.
96-97
Lepoint G, Nyssen F, Gobert S, Dauby P, Bouquegneau JM (2000) Relative impact of a seagrass bed and its adjacent epilithic algal community in consumer diets. Mar Biol 136:513-518.
192
VII. Synthesis
Synthesis
This study set out to describe energy flow and trophic relationships in two shallowwater benthic ecosystems in East Antarctica. By using stable isotopes of carbon and
nitrogen, I was able to identify the use of different carbon sources and trace the flow of
carbon from producers to consumers. This allowed me to build and test a descriptive
food web model that can be used as a benchmark against which future changes can be
compared. In addition to providing empirical data on food webs at the Windmill Islands
and Vestfold Hills, I assessed more broadly the state of the literature on marine food
webs that utilised isotopes of carbon and nitrogen. That analysis highlighted the role of
this body of literature in advancing food web theory.
My research emphasized both the uniqueness and similarity of Antarctic shallow-water
benthic communities to those from other parts of Antarctica and the rest of the world.
Nearshore food webs at the Windmill Islands and the Vestfold Hills share several
characteristics: (1) they rely on much the same carbon sources; (2) they have similar
food web heights and widths and reach a food chain length maxima equivalant to four
trophic levels (excluding marine birds and mammals); (3) they typify trophic continua,
rather than having discrete trophic levels; (4) higher-order consumers display high
levels of trophic omnivory amongst species; and (5) higher-order trophic guilds display
a lack of distinct trophic grouping.
Many of these food web properties are similar to food webs developed for benthic
communities on the Antarctic Peninsula (Kaehler et al. 2000, Dunton 2001, Corbisier et
al. 2004, Jacob et al. 2006), which also have four trophic levels and a high degree of
trophic omnivory among higher-orders. In my studies and those conducted on the
Antarctic Peninsula, pelagic POM and macroalgal POM constituted the major food
source for suspension feeders, whilst benthic and epiphytic diatoms were an additional
food source for grazers and surface-deposit feeders. This is in agreement with Kaehler
et al. (2000) and Dunton (2001) who estimated that between 25-68% of ingested
material is derived from macroalgae in food webs from Prince Edward Island and at
several
islands
near
Palmer
Station.
Corbisier
et
al.
(2004)
also
found
microphytobenthos and sediment detritus contributed to the diets of benthic grazers in a
food web from King George Island. Sea ice POM, despite its considerable biomass,
appears to provide only a limited direct input of carbon into the benthic system, except
193
VII. Synthesis
for those species which are able to migrate to the underside of sea ice during winter
months (e.g. Paramoera walkeri).
From these results it is evident that the food webs of the shallow-water benthic
communities of the Windmill Islands and Vestfold Hills are similar to those from the
islands and coastal regions of the Antarctic Peninsula. This may not come as a surprise:
whilst East Antarctica and the Antarctic Peninsula are separated by over 6000 km of
coastline and up to 15 degrees of latitude, there are few major ecological or physical
differences between both regions. Many of the flora and fauna that occupy shallow
waters on the Antarctic Peninsula also occur at higher latitudes (Dayton et al. 1994,
Gambi et al. 1994, Barnes et al. 2006, Gutt 2007). Both regions also experience strong
seasonality in primary production (Clarke 1988, Quartino and Boraso de Zaixso 2008)
and have similar disturbance regimes (Gambi et al. 1994, Barnes 1999, Nonato et al.
2000). Climatic conditions in terrestrial habitats may be milder on the Peninsula
compared to the continent, but differences in ocean temperature are much smaller,
varying less than 3°C between the Antarctic Peninsula and higher latitudes (Clarke
1988, Peck 2005).
Though I acknowledge the preliminary nature of this work, these findings are
nevertheless of considerable importance. The Antarctic Peninsula is undergoing rapid
warming due to climate change, in contrast to East Antarctica which has, so far,
remained relatively stable (Cook et al. 2005, Meredith and King 2005, Turner et al.
2005). Despite the uncertainty of when and how much East Antarctica is expected to
warm, it is likely that changes will be strongest and occur first, on the Peninsula. Any
changes to Antarctic Peninsula food webs can, therefore, act as an early warning of
what is likely to occur in East Antarctica. Food webs of East Antarctica can, in turn, act
as a baseline against which changes to food webs on the Peninsula can be assessed. Not
only can this occur with respect to climate change, but also other disturbance
mechanisms such as introduced species and impacts related to increasing human
occupation. Each of these disturbances are likely to affect benthic communities on the
Antarctic Peninsula before those at higher latitudes.
Of course, any location south of the Antarctic Peninsula could potentially act as a food
web baseline, and East Antarctica does not itself need to become the yardstick against
which other Antarctic food webs are compared. Rather, my research has shown that,
despite the distance and latitudinal separation between the Antarctic Peninsula and East
194
VII. Synthesis
Antarctica, and the wide geographical separation between the Windmill Islands and
Vestfold Hills in East Antarctica, food webs across these locations show similar
properties. It is therefore likely that other shallow-water locations around Antarctica
may show similar features.
From the analysis conducted in Chapter VI, we may conclude that shallow-water food
webs from the Antarctic Peninsula and East Antarctica are by no means extraordinary
compared to coastal food webs from temperate or tropical systems. They share similar
numbers of primary producers and consumers, food chain heights and ranges in δ13C.
What is likely to be fundamentally different are: the speed at which processes occur;
rates of carbon turn over; and the seasonality of primary production and growth. Hence,
superficially at least, Antarctic shallow-water benthic food webs resemble most other
shallow-water benthic food webs, yet the ‘engine’ driving these communities is likely to
be very different. These differences will be most pronounced when Antarctic
communities react to new disturbances such as climate change, which may have
pronounced effects on their resilience. Based on the research in this thesis, we may
conclude that Antarctic food webs have the same structural elements as most other
coastal marine food webs.
Future directions
Three streams of research need to be established to assess my hypothesis that circumAntarctic shallow-water benthic communities share very similar food web properties.
Firstly, the food web model requires further refinement. Whilst stable isotopes provide
an ideal first step in establishing the ‘frame work’ for food webs, this framework lacks
the detailed dietary information required to quantify the strength of feeding links.
Detailed feeding studies, using either traditional (gut contents) or new (gut DNA
analysis) methods, will provide detail for the model so that it more closely resembles
traditional food web models. Specifically this will facilitate description of food web
parameters such as link density and link strength, allowing the model to more closely
resemble a true energy web. Whilst this is a slow and tedious process, stable isotopes
can provide guidance in which species most require further dietary analysis, reducing
the need for widespread and more laborious dietary studies.
At both the Windmill Islands and Vestfold Hills, and indeed in most locations around
Antarctica, there is considerable work yet to be done to quantify species richness and
195
VII. Synthesis
biomass. Only the infaunal communities at the Windmill Islands have been described
quantitatively (Stark 2000), and only qualitative descriptions of epifaunal communities
are available for both locations (Dhargalkar et al. 1988, Kirkwood and Burton 1988,
Tucker and Burton 1988, Johnston et al. 2007, Clark et al. 2011). There have been no
biomass estimates of algae, or epifaunal communities, at either location. Quantitative
estimates will allow proper descriptions of energy flow, enabling determination of key
processes, species and feedback loops.
A second string of research should focus on understanding the importance of shallowwater communities in a regional context. We do not currently know whether shallowwater systems are essentially coastal islands, marooned amongst deeper waters, or
whether they are continuous with shelf communities. Nor do we understand the
contribution of shallow-water production to adjacent deep-water coastal areas in
providing additional food sources. If shallow-water coastal areas are as sparse as
estimated for continental ice-free areas (0.01%) (Snape et al. 2001), then they may
constitute rare or critical ecosystems which should be preserved in a system of reserves.
Likewise, if shallow-water areas provide considerable food resources for adjacent
regions (e.g. Reichardt 1987, Polis et al. 1997) then they constitute an important donor
area and should likewise be conserved.
We currently have insufficient data to assess either of these possibilities. For example
we do not understand what type of communities exist either side of the Windmill
Islands and Vestfold Hills, particularly where the continent meets the sea with little to
no shallow water. What occurs in these deeper areas? Do these communities resemble
shelf communities or are they continuations of shallow-water communities? Whilst
finding the answers to these questions requires considerable logistical and financial
support, we will not be able to classify the importance of the shallow-water
communities for which we have data, until we can clearly define their size, connectivity
and uniqueness relative to other regions of Antarctica.
Lastly, future research should seek to harness the increasing collection of global isotope
data. New dietary databases, such as those held at the Australian Antarctic Division
Data Centre (Raymond et al. 2011), should prove extremely useful in providing
researchers with detailed dietary information on many species, enhancing food web
descriptions. Other global databases, such as Barcode of Life or World Register of
Marine Species (WoRMS), are now valuable tools in the marine ecologist’s toolbox,
196
VII. Synthesis
and there is no reason why a trophic element (possibly summarised as an isotopic index)
could not be added. This would provide researchers with an extremely useful
standardised description of the trophic role of a species.
To ensure these measures can be successfully utilised, we must first confirm they are
both representative of the food web and are standardised. Chapter VI identified suitable
metrics from food web studies that utilised stable isotopes of carbon and nitrogen
(SIFWS) and provided the first steps in achieving the above goals. I was able to
successfully demonstrate that simple, common metrics from SIFWS can provide useful
information for cross-system comparisons. The next step in this research would be to
take a smaller subset of studies, all of which quantified sampling effort and reported
isotopic values in a similar fashion, in order to demonstrate that standardised reporting
of stable isotope can facilitate both weak and strong patterns across food webs from
different habitats or locations.
Future research must juggle the requirements of large-scale studies against the smallscale variability which can potentially confound interpretation of results at larger scales.
Chapter III highlighted the importance of documenting such variation, but also showed
the level of detail which could be obtained in studies that are mindful of local variation.
Accounting for localised variation should be considered when reporting isotopic
information, and be included in data standardisation protocols. This would provide
future researchers with confidence that comparative analyses of studies across systems
are not confounded by small-scale variation within systems.
All three avenues of research will provide data that are essential for objective
management of Antarctic coastal ecosystems into the future. On ground research
focusing on improving our knowledge of shallow-water food webs will provide
policymakers and managers from participating nations, and the Convention for the
Conservation of Antarctic Marine Living Resources (CCAMLR), with data on changing
ecosystems as a result of human disturbances, with the end result of better on-ground
management. Research geared towards understanding the connectivity of shallow-water
communities will provide decision makers with better information on priority areas for
conservation. Lastly, improved methods, facilitating cross-system comparison of food
webs, will not only provide additional information for the two streams of research
outlined above, but can also act as tools to advance food web theory.
197
VII. Synthesis
Limitations
The broad aim of this study was to advance our knowledge of shallow-water
communities from both the Windmill Islands and Vestfold Hills, two locations which
have undergone human occupation and study since the 1950s. Although several studies
have described species occurring at both locations, only one survey has provided a
detailed list of species from the Vestfold Hills (Tucker and Burton 1987): no “official”
list has been generated for the Windmill Islands (although work over the last 20 years
by the Department of Environmental Protection and Change, Australian Antarctic
Division (AAD), has resulted in a fauna list and reference collection in various stages of
identification). There is also currently no official field guide to the fauna occupying the
shallow waters of East Antarctica (although the AAD expects to release a guide in
2014). Consequently, morphospecies identification was necessary for several groups of
taxa, limiting much of our interpretation to commonly occurring species. A quantitative
survey of the benthic community and its structure is urgently needed at both locations,
and should be a high priority for future ecological work.
This study was also constrained by logistical considerations at both locations. Samples
from the Windmill Islands were either collected before the commencement of this
study, or in 2008 through surface methods, without the use of divers. Consequently, the
number of samples and species collected were limited. Likewise, for the Vestfold Hills,
the lack of previous ecological work restricted sample collection to qualitative,
‘representative’ samples of the community. On both occasions, sample collection was
restricted to a single opportunity: hence, missing data could not be recollected. This was
particularly crucial for benthic diatoms which were subsequently found to be a key
carbon resource for grazers and surface deposit feeders.
The exploratory nature of much of my research also meant that I was limited in
describing only the main features of the food web. I was unable to describe key detrital
pathways or feedback loops, which are essential components in any food web (Polis
1991, Polis and Strong 1996). Again, little research of this nature has been conducted
for any shallow-water system in Antarctica, and I acknowledge that both food webs in
my study describe only key consuming processes, up the food web, and fail to
adequately describe detrital pathways (down the food web). The addition of both detrital
processes and feedback loops should be included in future versions of my food web
198
VII. Synthesis
model. It is also pertinent to note that almost no work has been done to assess the
importance of competition at either location.
Both the limited nature of the ecological work which formed the foundation for my
research, and the restrictions placed on the study by field logistics, resulted in a study
that evolved over the last four years based on the time constraints imposed by getting to,
and working at, the Australian research stations. Despite these limitations, I was able to
clearly define the key processes in both food webs and develop a testable model
describing those key processes. Earlier versions of this work also led to the realisation
that there is a real need for comparative analysis amongst food webs and that stable
isotopes can be a useful tool to support standardisation across studies, facilitating
comparative analysis. These observations led to a review of the available food
web/isotope literature, and to the discovery of the relative lack of work undertaken in
this area. Chapter VI thus provides the first attempt to place Antarctic food webs into a
global context.
Concluding remarks
Shallow-water benthic food webs from East Antarctica are complex, yet there are
several, identifiable pathways for carbon to enter and move throughout the benthic
community. It is clear that a considerable amount of work still needs to be conducted
before we can gain a real understanding of the role of shallow-water benthic
communities in the wider coastal ecosystem. Currently, research in this area is
fragmented and extremely limited. Considering the likely importance of shallow-water
communities to adjacent communities and, at least in East Antarctica, the limited
distribution of shallow-water areas along the coast, the communities they contain will
likely need protection in the near future. Hence, it is imperative we act now to better
map and understand the roles of these shallow-water benthic communities.
Measurement of stable isotopes provides a critical first step in describing the framework
of food webs, and these methods are ideally suited for use in areas where samples are
extremely difficult to collect. Whilst acknowledging their limitations, analyses of stable
isotopes were successful in identifying trophic relationships amongst the benthic fauna
and were considered the best method under the limitations of this study, which was
constrained by the timelines for a PhD, logistics and finances. Despite these limitations,
the study highlighted the importance of Antarctic shallow-water benthic communities
199
VII. Synthesis
and provides a strong indication of their functional role in the wider Antarctic coastal
ecosystem. Importantly, the food web models, which are the primary outcome of the
study,
provide
a
clear
framework
for
200
future,
hypothesis-driven
research.
References
References
(Introduction & Synthesis)
B
Barnes D, Linse K, Waller C, Morely S, Enderlein P, Fraser K, Brown M (2006)
Shallow benthic fauna communities of South Georgia Island. Polar Biol 29:223228
Barnes DKA (1995) Sublittoral epifaunal communities at Signy Island, Antarctica II.
Below the ice foot zone. Mar Biol 121:565-572
Barnes DKA (1999) The influence of ice on the polar nearshore benthos. J Mar Biol
Assoc UK 79:401-407
Blight D, Oliver R (1977) The metamorphic geology of the Windmill Islands,
Antarctica: a preliminary account. Aust J Earth Sci 24:239-262
Bokhorst S, Ronfort C, Huiskes A, Convey P, Aerts R (2007) Food choice of Antarctic
soil arthropods clarified by stable isotope signatures. Polar Biol 30:983-990
Bouillon S, Raman AV, Dauby P, Dehairs F (2002) Carbon and nitrogen stable isotope
ratios of subtidal benthic invertebrates in an estuarine mangrove ecosystem
(Andhra Pradesh, India). Estuar Coast Shelf Sci 54:901-913
Brenner M, Buck B, Cordes S, Dietrich L, Jacob U, Mintenbeck K, Schröder A, Brey T,
Knust R, Arntz W (2001) The role of iceberg scours in niche separation within
the Antarctic fish genus Trematomus. Polar Biol 24:502-507
C
Cabana G, Rasmussen JB (1996) Comparison of aquatic food chains using nitrogen
isotopes. Proc Natl Acad Sci USA 93:10844-10847
Caquet T (2006) Use of carbon and nitrogen stable isotope ratios to assess the effects of
environmental contaminants on aquatic food webs. Environ Pollut 141:54-59
Carlier A, Riera P, Amouroux J-M, Bodiou J-Y, Gremare A (2007) Benthic trophic
network in the Bay of Banyuls-sur-Mer (northwest Mediterranean, France): an
201
References
assessment based on stable carbon and nitrogen isotopes analysis. Estuar Coast
Shelf Sci 72:1-15
Chikaraishi Y, Ogawa N, Kashiyama Y, Takano Y, Suga H, Tomitani A, Miyashita H,
Kitazato H, Ohkouchi N (2009) Determination of aquatic food-web structure
based on compound-specific nitrogen isotopic composition of amino acids.
Limnol Oceanogr Meth 7:740-750
Clark G, Stark J, Perrett L, Hill N, Johnston E (2011) Algal canopy as a proxy for the
disturbance history of understorey communities in East Antarctica. Polar Biol
34:781-790
Clarke A (1985) Energy flow in the Southern Ocean food web. In: Siegfried WR,
Condy PR, Laws RM (eds) Antarctic nutrient cycles and food webs. SpringerVerlag, Berlin Heidelberg
Clarke A (1988) Seasonality in the Antarctic marine environment. Comp Biochem
Physiol B 90:461-473
Cloern JE, Canuel EH, Harris D (2002) Stable carbon and nitrogen isotope composition
of aquatic and terrestrial plants of the San Francisco Bay estuarine system.
Limnol Oceanogr 47:713-729
Cohen JE, Beaver RA, Cousins SH, Deangelis DL, Goldwasser L, Heong KL, Holt RD,
Kohn AJ, Lawton JH, Martinez N, O'malley R, Page LM, Patten BC, Pimm SL,
Polis GA, Rejmãnek M, Schoener TW, Schoenly K, Sprules WG, Teal JM,
Ulanowicz RE, Warren PH, Wilbur HM, Yodzis P (1993) Improving food webs.
Ecology 74:252-258
Connolly RM (2003) Differences in trophodynamics of commercially important fish
between artificial waterways and natural coastal wetlands. Estuar Coast Shelf
Sci 58:929-936
Conway NM, Kennicutt Ii MC, Van Dover CL (1994) Stable isotopes in the study of
marine chemosynthetic-based ecosystems. In: Lajtha K, Michener R (eds) Stable
isotopes in ecology and environmental science. Blackwell scientifc publications,
London
Cook AJ, Fox AJ, Vaughan DG, Ferrigno JG (2005) Retreating glacier fronts on the
Antarctic Peninsula over the past half-century. Science 308:541-544
202
References
Corbisier T, Petti MV, Skowronski RP, Brito TS (2004) Trophic relationships in the
nearshore zone of Martel Inlet (King George Island, Antarctica): δ13C stableisotope analysis. Polar Biol 27:75-82
Costanzo SD, O'donohue MJ, Dennison WC, Loneragan NR, Thomas M (2001) A new
approach for detecting and mapping sewage impacts. Mar Pollut Bull 42:149156
D
Dayton PK, Gordon AR, Devries AL (1969) Anchor ice formation in McMurdo Sound,
Antarctica, and its biological effects. Science 163:273-274
Dayton PK, Gordon AR, Paine RT, Dayton LB (1974) Biological accommodation in the
benthic community at McMurdo Sound, Antarctica. Ecol Monogr 44:105-128
Dayton PK, Mordida BJ, Bacon F (1994) Polar marine communities. Am Zool 34:90-99
Dayton PK, Oliver JS (1977) Antarctic soft-bottom benthos in oligotrophic and
eutrophic environments. Science 197:55-58
Deniro MJ, Epstein S (1978) Influence of diet on the distribution of carbon isotopes in
animals. Geochim Cosmochim Acta 42:495-506
Deniro MJ, Epstein S (1981) Influence of diet on the distribution of nitrogen isotopes in
animals. Geochim Cosmochim Acta 45:341-351
Dhargalkar VK (1990) Benthic marine algae of the inshore water at the Vestfold Hills,
Antarctica. Indian J Mar Sci 19:110-114
Dhargalkar VK, Burton HR, Kirkwood JM (1988) Animal associations with the
dominant species of shallow water macrophytes along the coastline of the
Vestfold Hills, Antarctica. Hydrobiologia 165:141-150
Ducklow HW, Fraser W, Karl DM, Quetin LB, Ross RM, Smith RC, Stammerjohn SE,
Vernet M, Daniels RM (2006) Water-column processes in the West Antarctic
Peninsula and the Ross Sea: interannual variations and food web structure. Deep
Sea Res Pt II 53:834-852
Dunton KH (2001) δ15N and δ13C measurements of Antarctic Peninsula fauna: trophic
relationships and assimilation of benthic seaweeds. Amer Zool 41:99-112
203
References
Dunshea G (2009) DNA-based diet analysis for any predator. PLoS ONE 4:e5252
E
Echeverría C, Paiva P (2006) Macrofaunal shallow benthic communities along a
discontinuous annual cycle at Admiralty Bay, King George Island, Antarctica.
Polar Biol 29:263-269
Ehleringer JR, Rundel PW (1989) Stable isotopes: history, units and instrumentation.
In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable isotopes in ecological
research, book 68. Springer-Verlag, New York
Everitt D, Poore G, Pickard J (1980) Marine benthos from Davis Station, Eastern
Australia. Mar Fresh Res 31:829-836
F
Fabry VJ, Seibel BA, Feely RA, Orr JC (2008) Impacts of ocean acidification on marine
fauna and ecosystem processes. ICES J Mar Sci 65:414-432
Fantle MS, Dittel AI, Schwalm SM, Epifanio CE, Fogel ML (1999) A food web
analysis of the juvenile blue crab, Callinectes sapidus, using stable isotopes in
whole animals and individual amino acids. Oecologia 120:416-426
France RL (1995) Carbon-13 enrichment in benthic compared to planktonic algae:
foodweb implications. Mar Ecol Prog Ser 124:307-312
Fry B (1988) Food web structure on Georges Bank from stable C, N, and S isotopic
compositions. Limnol Oceanogr 33:1182-1190
Fry B (2006) Stable isotope ecology, Vol. Springer, New York
Fry B, Baltz DM, Benfield MC, Fleeger JW, Gace A, Hass HL, Quinones-Rivera Z
(2003) Stable isotope indicators of movement and residency for brown shrimp
(Farfantepenaeus aztecus) in coastal Louisiana marshscapes. Estuaries 26:82-97
Fry B, Sherr EB (1989) 13C measurments as indicators of carbon flow in marine and
freshwater ecosystems. In: Rundel PW, Ehleringer JR, Nagy KA (eds) Stable
isotopes in ecological research, book 68. Springer-Verlag, New York
204
References
G
Gambi MC, Lorenti M, Russo GF, Scipione MB (1994) Benthic associations of the
shallow hard bottoms off Terra Nova Bay, Ross Sea: zonation, biomass and
population structure. Antarct Sci 6:449-462
Gannes LZ, Del Rio CM, Koch P (1998) Natural abundance variations in stable isotopes
and their potential uses in animal physiological ecology. Comp Biochem Physiol
A 119:725-737
Gaye-Siessegger J, Focken U, Abel H, Becker K (2007) Starvation and low feeding
levels result in an enrichment of
13
C in lipids and
15
N in protein of Nile tilapia
Oreochromis niloticus. J Fish Biol 71:90-100
Gaye-Siessegger J, Focken U, Muetzel S, Abel H, Becker K (2004) Feeding level and
individual metabolic rate affect δ13C and δ15N values in carp: implications for
food web studies. Oecologia 138:175-183
Gervais F, Riebesell U (2001) Effect of phosphorus limitation on elemental composition
and stable carbon isotope fractionation in a marine diatom growing under
different CO2 concentrations. Limnol Oceanogr 46:497-504
Goericke R, Montoya JP, Fry B (1994) Physiology of isotope fractionation in alage and
cyanobacteria. In: Lajtha K, Michener R (eds) Stable isotopes in ecology and
environmental science. Blackwell scientific publications, London
Grebmeier JM, Barry JP (1991) The influence of oceanographic processes on pelagicbenthic coupling in polar regions: a benthic perspective. J Mar Syst 2:495-518
Griffiths H (1991) Applications of stable isotope technology in physiological ecology.
Funct Ecol 5:254-269
Gutt J (2000) Some "driving forces" structuring communities of the sublittoral Antarctic
macrobenthos Antarct Sci 12:297-313
Gutt J (2007) Antarctic macro-zoobenthic communities: a review and an ecological
classification. Antarct Sci 19:165-182
Gutt J, Koubbi P, Eléaume M (2007) Mega-epibenthic diversity off Terre Adélie
(Antarctica) in relation to disturbance. Polar Biol 30:1323-1329
205
References
Gutt J, Starmans A (2003) Patchiness of the megabenthos at small scales: ecological
conclusions by examples from polar shelves. Polar Biol 26:276-278
H
Hadwen WL, Arthington AH (2007) Food webs of two intermittently open estuaries
receiving 15N-enriched sewage effluent. Estuar Coast Shelf Sci 71:347-358
Hadwen WL, Russell GL, Arthington AH (2007) Gut content- and stable isotopederived diets of four commercially and recreationally important fish species in
two intermittently open estuaries. Mar Fresh Res 58:363-375
Hempel G (1985) Antarctic marine food webs. In: Siegfried WR, Condy PR, Laws RM
(eds) Antarctic nutrient cycles and food webs. Springer-Verlag, Berlin
Heidelberg
Hobson KA (1999) Tracing origins and migration of wildlife using stable isotopes: a
review. Oecologia 120:314-326
Hobson KA, Clark RG (1992) Assessing avian diets using stable isotopes II: factors
influencing diet-tissue fractionation. The Condor 94:189-197
Hood-Nowotny R, Knols BGJ (2007) Stable isotope methods in biological and
ecological studies of arthropods. Entomol Exp Appl 124:3-16
Hopkins TL, Ainley DG, Torres JJ, Lancraft TM (1993) Trophic structure in open
waters of the marginal ice zone in the Scotia-Weddell confluence region during
spring (1983). Polar Biol 13:389-397
I
Iken K, Barrera-Oro ER, Quartino ML, Casaux RJ, Brey T (2004) Grazing by the
Antarctic fish Notothenia coriiceps: evidence for selective feeding on
macroalgae. Antarct Sci 9:386-391
J
Jacob U, Brey T, Fetzer I, Kaehler S, Mintenbeck K, Dunton K, Beyer K, Struck U,
Pakhomov EA, Arntz WE (2006) Towards the trophic structure of the Bouvet
Island marine ecosystem. Polar Biol 29:106-113
206
References
Jarman SN, Deagle BE, Gales NJ (2004) Group-specific polymerase chain reaction for
DNA-based analysis of species diversity and identity in dietary samples. Mol
Ecol 13:1313-1322
Jarre-Teichmann A, Brey T, Bathman U, Dahm C, Dieckmann G, Gorny M, Klages M,
Pages F, Plotz J, Schnack-Schiel SB, Stiller M, Arntz W (1997) Trophic flows in
the benthic shelf community of the eastern Weddell Sea, Antarctica. In:
Antarctic communities: species, structure and survival. Cambridge University
Press, Cambridge
Johnston E, Connell S, Irving A, Pile A, Gillanders B (2007) Antarctic patterns of
shallow subtidal habitat and inhabitants in Wilke’s Land. Polar Biol 30:781-788
K
Kaehler S, Pakhomov EA, Mcquaid CD (2000) Trophic structure of the marine food
web at the Prince Edward Island (Southern Ocean) determined by δ13C and δ15N
analysis. Mar Ecol Prog Ser 208:13-20
Kelly JF (2000) Stable isotopes of carbon and nitrogen in the study of avian and
mammalian trophic ecology. Can J Zool 78:1-27
Kirkwood JM, Burton HR (1988) Macrobenthic species assemblages in Ellis Fjord,
Vestfold Hills, Antarctica. Mar Biol 97:445-457
Kling GW, Fry B, O'brian WJ (1992) Stable isotopes and planktonic trophic structure in
Arctic lakes. Ecology 73:561-566
Klöser H, Quartino ML, Wiencke C (1996) Distribution of macroalgae and macroalgal
communities in gradients of physical conditions in Potter Cove, King George
Island, Antarctica. Hydrobiologia 333:1-17
Knust R, Arntz WE, Boche M, Brey T, Gerdes D, Mintenbeck K, Schroder A, Starmans
A, Teixidó N (2003) Iceburg scouring on the Eastern Weddell Sea Shelf
(Antarctica): a benthic system shaped by physical disturbances? In: Huiskes
AHL, Gieskes WWC, Rozema J, Schorno RML, van der Vies SM, Wolff WJ
(eds) Antarctic biology in a global context. Backhuys Publishers, Leiden, The
Netherlands
207
References
L
Lajtha K, Marshall JD (1994) Sources of variation in the stable isotopic composition of
plants. In: Lajtha K, Michener R (eds) Stable isotopes in ecology and
environmental science. Blackwell scientific publications, London
Leibold MA, Chase JM, Shurin JB, Downing AL (1997) Species turnover and the
regulation of trophic structure. Annu Rev Ecol Syst 28:467-494
M
Maberly SC, Raven JA, Johnston AM (1992) Discrimination between
12
C and
13
C by
marine plants. Oecologia 91:481-492
Massom RA, Stammerjohn SE (2010) Antarctic sea ice change and variability physical and ecological implications. Polar Science 4:149-186
Matthews B, Mazumder A (2004) A critical evaluation of intrapopulation variation of
δ13C and isotopic evidence of individual specialization. Oecologia 140:361-371
May RM (1979) Ecological interactions in the Southern Ocean. Nature 277:86-89
McClelland JW, Montoya JP (2002) Trophic relationships and the nitrogen isotopic
composition of amino acids in plankton. Ecology 83:2173-2180
McConnaughey T, Mcroy CP (1979) Food-web structure and the fractionation of
Carbon isotopes in the bering sea. Mar Biol 53:257-262
McMahon KW, Ambrose Jr WG, Johnson BJ, Sun M-Y, Lopez GR, Clough LM,
Carroll ML (2006) Benthic community response to ice algae and phytoplankton
in Ny Ålesund, Svalbard. Mar Ecol Prog Ser 310:1-14
McMinn A, Runcie JW, Riddle MJ (2004) Effects of seasonal sea ice breakout on the
photosynthesis of benthic diatom mates at Casey, Antarctica. J Phycol 40:62-69
Meredith MP, King JC (2005) Rapid climate change in the ocean west of the Antarctic
Peninsula during the second half of the 20th century. Geophys Res Lett 32
Michener R, Schell DM (1994) Stable isotopes as tracers in marine aquatic food webs.
In: Lajtha K, Michener R (eds) Stable isotopes in ecology and environmental
science. Blackwell scientific publications, London
208
References
Minagawa M, Wada E (1984) Stepwise enrichment of
15
N along food chains: further
evidence and the relation between δ15N and animal age. Geochim Cosmochim
Acta 48:1135-1140
Mincks SL, Smith CR, Demaster DJ (2005) Persistence of labile organic matter and
microbial biomass in Antarctic shelf sediments: evidence of a sediment 'food
bank'. Mar Ecol Prog Ser 300:3-19
Murray MD, Luders DJ (1990) Faunistic studies at the Windmill Islands, Wilkes Land,
East Antarctica, 1959-80. ANARE Research Notes 73:45pp
N
Nonato EF, Brito TAS, De Paiva PC, Petti MAV, Corbisier TN (2000) Benthic
megafauna of the nearshore zone of Martel Inlet (King George Island, South
Shetland Islands, Antarctica): depth zonation and underwater observations. Polar
Biol 23:580-588
Norkko A, Thrush SF, Cummings V, Gibbs HL, Andrew N, Norkko J, Schwarz A-M
(2007) Trophic structure of coastal Antarctic food webs associated with changes
in sea ice and food supply Ecology 88:2810-2820
Nyssen F, Brey T, Lepoint G, Bouquegneau J-M, De Broyer C, Dauby P (2002) A
stable isotope approach to the eastern Weddell Sea trophic web: focus on
benthic amphipods. Polar Biol 25:280-287
O
Oliver JS, Slattery PN (1985) Effects of crustacean predators on species composition
and population structure of soft-bodied infauna from McMurdo Sound,
Antarctica. Ophelia 24:155-175
Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, Feely RA, Gnanadesikan A, Gruber
N, Ishida A, Joos F, Key RM, Lindsay K, Maier-Reimer E, Matear R, Monfray
P, Mouchet A, Najjar RG, Plattner G-K, Rodgers KB, Sabine CL, Sarmiento JL,
Schlitzer R, Slater RD, Totterdell IJ, Weirig M-F, Yamanaka Y, Yool A (2005)
209
References
Anthropogenic ocean acidification over the twenty-first century and its impact
on calcifying organisms. Nature 437:681-686
P
Pasquaud S, Lobry J, Elie P (2007) Facing the necessity of describing estuarine
ecosystems: a review of food web ecology study techniques. Hydrobiologia
588:159-172
Peck L (2005) Prospects for surviving climate change in Antarctic aquatic species.
Front Zool 2:9
Peterson BJ (1999) Stable isotopes as tracers of organic matter input and transfer in
benthic food webs: a review. Acta Oecol 20:479-487
Peterson BJ, Fry B (1987) Stable isotopes in ecosystem studies. Annu Rev Ecol Syst
18:293-320
Pimm SL (1982) Food webs, Vol. Chapman and Hall, New York
Pimm SL (1984) The complexity and stability of ecosystems. Nature 307:321-325
Polis GA (1991) Complex trophic interactions in deserts: an empirical critique of foodweb theory. Amer Nat 138:123-155
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
Polis GA, Strong DR (1996) Food web complexity and community dynamics. Amer
Nat 147:813-846
Post DM (2002) The long and short of food-chain length. Trends Ecol Evol 17:269-277
Q
Quartino M, Boraso De Zaixso A (2008) Summer macroalgal biomass in Potter Cove,
South Shetland Islands, Antarctica: its production and flux to the ecosystem.
Polar Biol 31:281-294
210
References
R
Rau GH, Hopkins TL, Torres JJ (1991)
15
N/14N and
13
C/12C in Weddell Sea
invertebrates: implications for feeding diversity. Mar Ecol Prog Ser 77:1-6
Rau GH, Sweeney RE, Kaplan IR (1982) Plankton 13C:12C ratio changes with latitude:
differences between northern and southern oceans. Deep Sea Res Pt I 29:10351039
Raymond B, Marshall M, Nevitt G, Gillies CL, Van Den Hoff J, Stark JS, Losekoot M,
Woehler EJ, Constable AJ (2011) A Southern Ocean dietary database. Ecology
92:1188-1188
Reichardt WT (1987) Burial of Antarctic macroalgal debris in bioturbated deep-sea
sediments. Deep Sea Res Pt I 34:1761-1770
Rounick JS, Winterbourn MJ (1986) Stable carbon isotopes and carbon flow in
ecosystems. Bioscience 36:171-177
Rudnick D, Resh V (2005) Stable isotopes, mesocosms and gut content analysis
demonstrate trophic differences in two invasive decapod crustacea. Freshwat
Biol 50:1323-1336
S
Sahade R, Tatián M, Kowalke J, Kühne S, Esnal GB (1998) Benthic faunal associations
on soft substrates at Potter Cove, King George Island, Antarctica. Polar Biol
19:85-91
Schoener TW (1989) Food webs from the small to the large: the Robert H. MacArthur
award lecture. Ecology 70:1559-1589
Shearer G, Kohl DH (1989) Estimates of N2 fixation in ecosystems: the need for and
basis of the 15N natural abundance method In: Rundel PW, Ehleringer JR, Nagy
KA (eds) Stable isotopes in ecological research, Book 68. Springer-Verlag, New
York
Siegfried WR, Condy PR, Laws RM (1985) Antarctic nutrient cycles and food webs,
Vol. Springer-Verlag, Berlin Heidelberg
211
References
Smith CR, Mincks S, Demaster DJ (2006) A synthesis of bentho-pelagic coupling on
the Antarctic shelf: food banks, ecosystem inertia and global climate change.
Deep Sea Res Pt II 53:875-894
Snape I, Riddle MJ, Stark JS, King CK, Duquesne S, Gore DB (2001) Managment and
remediation of contaminated sites at Casey Station, Antarctica. Polar Rec
37:199-214
Stark JS (2000) The distribution and abundance of soft-sediment macrobenthos around
Casey Station, East Antarctica. Polar Biol 23:840-850
Stark JS, Riddle MJ, Simpson RD (2003) Human impacts in soft-sediment assembalges
at Casey Station, East Antarctica: spatial variation, taxonomic resolution and
data transformation. Austral Ecol 28:287-304
Sweeney RE, Kaplan IR (1980) Tracing flocculent industrial and domestic sewage
transport on San Pedro Shelf, Southern California, by nitrogen and sulphur
isotope ratios. Mar Environ Res 3:215-224
T
Tatián M, Sahade R, Mercuri G, Fuentes V, Antacli J, Stellfeldt A, Esnal G (2008)
Feeding ecology of benthic filter-feeders at Potter Cove, an Antarctic coastal
ecosystem. Polar Biol 31:509-517
Tieszen LL, Boutton TW, Tesdahl KG, Slade NA (1983) Fractionation and turnover of
stable carbon isotopes in animal tissues: Implications for δ13C analysis of diet.
Oecologia 57:32-37
Tucker MJ, Burton HR (1987) A survey of the marine fauna in shallow coastal waters
of the Vestfold Hills and Rauer Islands, Antarctica. ANARE Research Notes
55:1-27
Tucker MJ, Burton HR (1988) The inshore marine ecosystem off the Vestfold Hills,
Antarctica. Hydrobiologia 165:129-139
Turner J, Colwell SR, Marshall GJ, Lachlan-Cope TA, Carleton AM, Jones PD, Lagun
V, Reid PA, Iagovkina S (2005) Antarctic climate change over the last 50 years.
Int J Climatol 25:279-294
212
References
V
Vander Zanden MJ, Fetzer WW (2007) Global patterns of aquatic food chain length.
Oikos 116:1378-1388
W
Wada E, Mizutani H, Minagawa M (1991) The use of stable isotopes for food web
analysis. Crit Rev Food Sci Nutr 30:361-371
Wada E, Terazaki M, Kabaya Y, Nemoto T (1987) δ15N and δ13C abundances in the
Antarctic Ocean with emphasis on the biogeochemical structure of the food web.
Deep Sea Res Pt I 34:829-841
Walter DE, Kaplan DT, Permar TA (1991) Missing links: a review of methods used to
estimate trophic links in soil food webs. Agric, Ecosyst Environ 34:399-405
Williams R (1988) The inshore marine fishes of the Vestfold Hills region, Antarctica.
Hydrobiologia 165:161-167
Wong WW, Sackett WM (1978) Fractionation of stable carbon isotopes by marine
phytoplankton. Geochim Cosmochim Acta 42:1809-181
213
Appendicies
Appendices
Reprints of published papers
214