© The Author 2017. Published by Oxford University Press on behalf of
the International Society for Behavioral Ecology. All rights reserved. For
permissions, please e-mail: email@example.com
The ocial journal of the
International Society for Behavioral Ecology
Macronutrient selection of free-ranging urban
Australian white ibis (Threskiornis moluccus)
Sean C.P.Coogan,a,b Gabriel E.Machovsky-Capuska,a,b Alistair M.Senior,b,c
John M.Martin,d Richard E.Major,e and DavidRaubenheimera,b
aSchool of Life and Environmental Sciences, University of Sydney, Sydney, Australia, bCharles Perkins
Centre, University of Sydney, Sydney, Australia, cSchool of Mathematics and Statistics, University of
Sydney, Sydney, Australia, dRoyal Botanic Gardens and Domain Trust, Sydney, Australia, and eAustralian
Museum Research Institute, Australian Museum, Sydney, NSW, Australia
Received 24 November 2016; revised 11 March 2017; editorial decision 23 March 2017; accepted 29 March 2017.
Anthropogenic habitats often provide urban wildlife the opportunity to feed on a range of nutritionally diverse foods, which may ulti-
mately lead to human-wildlife conﬂict. The Australian white ibis (Threskiornis moluccus) provides an exemplar model for examining the
nutritional priorities and constraints of a native vertebrate that is successfully transitioning to an urban specialist. Here, we used ﬁeld-
based feeding trials to investigate the macronutrient preferences of free-ranging ibis in Sydney, Australia. Feeding trials (n=61) offer-
ing three experimental feeds showed that ibis selected signiﬁcantly more high-carbohydrate (HC) than high-protein (HP) and high-lipid
(HL) foods (95% CIHP-HC=−1.115 to −0.709; CIHL-HC=−1.874 to −1.468), and signiﬁcantly more HP than HL (CIHL-HP=−0.962 to −0.556). The
average proportion of macronutrient-derived energy selected by ibis was 25% protein (P; ± 1.1 SE): 23% lipid (L; ± 1.1): 52% carbohy-
drate (C; ± 1.8). Nutritional geometry suggested that mixtures selected at experimental feeders were substantially higher in C and lower
in P than are natural prey (e.g. insects, crustaceans), which were composed primarily of P and L.Compositional log-ratio-based linear
models of factors affecting macronutrient proportions selected by ibis showed that: 1)ln(P/C) increased with amount of recent rainfall;
and 2)ln(L/C) also increased with rain and had a non-linear relationship with number of birds feeding. Our results suggest that ibis for-
age for macronutrients rather than energy “per se”, and that their urban foraging is inﬂuenced by competition and the environment.
Key words: carbohydrate, foraging, macronutrient selection, nutritional ecology, urban ecology, urban exploiter.
Urbanization has been identiﬁed as a major threat to global biodi-
versity, and its eects are expected to increase considerably owing
to the projected growth of the human population and urban areas
(Secr. Conv. Biol. Div. 2012; Seto etal. 2012; Aronson etal. 2014;
U. N.Dept. Econ. Soc. A. Pop. Div. 2015). Anthropogenic pres-
sures can present several challenges to wildlife inhabiting urban
areas (McKinney 2002; Kalnay and Cai 2003; Lowry etal. 2013).
They can also present opportunities, as evidenced by the many
non-human species that successfully inhabit anthropogenic envi-
ronments (McKinney 2002; McDonnell and Hahs 2015). For
example, wildlife occupying urban habitats may capitalize on
altered distributions, availabilities, and abundances of food relative
to “natural” habitats (Chace and Walsh 2006; Gordon etal. 2016).
The year-round availability of concentrated anthropogenic food
resources may ultimately aect aspects of the biology of urban
wildlife, potentially leading to changes in traits such as migratory
behaviour and reproduction (Beckmann and Berger 2003; Martin
etal. 2011; Gilbert etal. 2016). Furthermore, urban food resources
may lead to incidences of human-wildlife conﬂict (Murray et al.
2015; Coogan and Raubenheimer 2016).
The Australian white ibis (hereafter “ibis”, Threskiornis moluccus)
is a native Australian wading bird whose distribution has recently
shifted from natural wetlands towards urban habitats (Martin
et al. 2010; Smith et al. 2013). The species therefore provides an
exemplar model for examining the nutritional priorities and con-
straints of a native vertebrate that is successfully transitioning to
an urban specialist. Traditionally, the ibis was found in large popu-
lations throughout inland wetlands in eastern Australia, and was
rarely found in urban areas (Carrick 1962; Cowling and Lowe
198l; Kingsford and Johnson 1998). Urban populations of ibis in
eastern Australia, however, have increased dramatically since the
1970s, likely in response to increased agricultural water extraction
from inland rivers, drought, habitat loss, and crucially their ability
to utilize urban food resources (Martin etal. 2007, 2010, 2012). In
Address correspondence to S.C.P. Coogan, The University of Sydney,
Charles Perkins Centre, D17, Level 4 East, NSW, 2006, E-mail: sean.
Behavioral Ecology (2017), 00(00), 1–9. doi:10.1093/beheco/arx060
fact, urban ibis can be considered pests and are involved in inci-
dences of human-wildlife conﬂict due to their habit of scavenging
anthropogenic foods from a variety of sources, including garbage
dumps, rubbish bins, and picnic tables (Ross 2004; Epstein et al.
2006; Martin etal. 2007, 2010).
Relatively few studies have investigated the natural diet of ibis,
although those which have, documented them as carnivores con-
suming a wide range of insect and small invertebrate and ver-
tebrate prey (Carrick 1959; Vestjens 1973; Barker and Vestjens
1990; Smith and Munro 2008). Interestingly, ibis in urban parks
have been shown to alternate foraging patterns depending on
previous rainfall. On days following rainfall, ibis were associ-
ated with parks with higher relative abundance of earthworms
(Oligochaeta), which they also consumed at a greater rate than
during dry periods (Chard 2015). Conversely, during dry peri-
ods ibis were associated with parks having higher availability of
anthropogenic foods (Chard 2015).
Despite foraging being cited as a main driver of ibis-human
conﬂicts, and the ability of ibis to forage on anthropogenic foods
being suggested as a key to their success in urban areas (Ross 2004;
Martin et al. 2007, 2010, 2011; Murray and Shaw 2009; Smith
2009), no studies have yet investigated the nutritional preferences
of the species. In other species, the ratios and amounts of protein,
lipid, and carbohydrate (and other nutrients) in foods have been
demonstrated to be key drivers of foraging behaviour and ﬁtness,
independent of energy intake “per se” (reviewed in Simpson and
Raubenheimer 2012; Nie etal. 2014; Raubenheimer et al. 2015).
Nutrient-speciﬁc foraging behaviour is likely to play a role in the
food choice of even obligate carnivores, which might face relatively
less variation in food composition than herbivores or omnivores
(Jensen etal. 2012; Kohl etal. 2015). The nutritional compositions
of foods found in anthropogenic habitats may be fundamentally
dierent than foods from the “native” range. Compared to non-
urban areas, carbohydrate and lipid are likely relatively more avail-
able than protein in an urban setting (Coogan and Raubenheimer
2016). Additionally, the macronutrient preferences of a species
may play a prominent role in food-related human-wildlife conﬂict
(Coogan and Raubenheimer 2016). Understanding the macronu-
trient-speciﬁc foraging behaviour of urban ibis may thus provide
valuable insight into their successful urban colonization and food-
Here, we combine ﬁeld-based experimental feeding trials, state-
space nutritional models (nutritional geometry), and compositional
data analysis to investigate the macronutrient preferences and
nutritional ecology of free-ranging ibis in an urban setting. We
test the prediction that ibis compose diets with non-random pro-
portions of macronutrients by addressing three questions. First,
do ibis select isoenergetic (i.e. equal energy density) foods experi-
mentally manipulated to contain predominantly one macronutri-
ent, high-protein (HP), high-lipid (HL), or high-carbohydrate (HC),
in unequal proportions (i.e. non-randomly)? Second, if ibis select
macronutrient proportions non-randomly, do they select primarily
HP and HL foods typical of a carnivorous diet (Kohl etal. 2015).
Third, is the selection of macronutrient ratios inﬂuenced by envi-
ronmental factors, such as the degree of intraspeciﬁc competition
and the amount of recent rainfall?
MATERIALS AND METHODS
This study was approved by the University of Sydney Animal
Ethics Committee (2015/927).
We oered ibis three pelleted feeds (HP, HL, and HC) as
described in Machovsky-Capuska et al. (2016a), with a slightly
modiﬁed formulation to improve the structural integrity of pellets
(Supplementary Table S1). The three dierent feeds were isoener-
getic (3.1 kcal/g) to remove the confound of ibis potentially select-
ing higher-energy-density pellets, composed of primarily naturally
derived rather than semi-synthetic ingredients, and balanced in
micro- and macro-minerals. Indigestible powdered cellulose was
used as a ﬁller to keep the energy content of foods constant. On a
wet-weight basis, HP pellets contained 48.0% crude protein (CP),
8.0% crude fat (CF), and 5.0% starch. HL pellets contained 2.0%
CP, 36.4% CF, and 0.1% starch. HC pellets contained 10.3% CP,
3.5% CF, and 58.2% starch.
Feeding stations and food consumption
Ibis in urban habitats are often associated with discrete anthropo-
genic features such as parks, landﬁlls, and ponds or wetlands, and
show high site ﬁdelity (Murray and Shaw 2009; Martin etal. 2011,
2012). The macronutrient selection study was conducted at Hyde
Park (~33°52′19″S, 151°12′37″E) and the Royal Botanic Gardens
(~33°51′46″S, 151°12′55″E; RBG), Sydney, New South Wales,
Australia. These areas were selected because they contain a rela-
tively large and consistent population of ibis as determined from
long term population surveys (Martin etal. 2010). The study was
conducted from April to June 2016, which was within the typical
non-breeding season (January to June) as deﬁned by Martin et al.
(2010). All feeding trials were held in the morning (i.e. between sun-
rise and 12:00 PM). Typically two feeding trials were performed
per morning, one in each park. Care was taken to ensure no birds
were fed more than once per day. We performed feeding trials dur-
ing rain-free periods to avoid feeds getting wet and to standardize
our sampling design.
Experimental foods were supplied in a dish intended to emulate
those used by visitors to the parks to store foods, for example, for
lunches and picnicking. This approach was taken because a pilot
study (undertaken during January to March 2016)revealed that ibis
were generally wary of unfamiliar or opaque dishes (e.g. those used
in Machovsky-Capuska etal. 2016a), and/or would not recognize
the feeding stations as sources of food, which may be consistent
with previous observations of urban birds being bold yet neophobic
(Audet etal. 2016). We used a 1.1L semi-transparent plastic food
storage container (18×11×6cm), with two semi-transparent plas-
tic dividers inserted in the dish to create three equal sized compart-
ments (Figure1). Each compartment was ﬁlled with 50g of one of
the three experimental foods (150g total). The type of food in the
middle position of the dish was alternated each experimental day
to account for positional eects. To prevent ibis from tipping the
dish over, we attached dulled clear-plastic disposable picnic knives
to the bottom of the dish (oriented across the width of the dish)
using clear packing tape to act as stabilizers.
After locating ibis (individuals or groups) in the park, we pre-
sented the food dish to the bird(s) and placed it on the ground to
allow them to feed. If birds initiated feeding, sessions were limited
to a maximum of 20 min to prevent large groups of birds from
potentially consuming all the foods. Often during feeding ses-
sions, additional ibis recruited to the feeding dish. Ibis typically
fed one at a time at the feeding dish, during which time they had
full access to all foods. Typically, one dominant individual in the
group would monopolize and defend the dish. Once a dominant
Page 2 of 9
Coogan etal. • Macronutrient selection of urbanibis
bird was ﬁnished feeding it would typically leave and another bird
would take its place, and so on. We recorded the total number of
birds observed feeding from the dish, including ibis that quickly
and seemingly indiscriminately rushed in to feed from the dish in
the presence of the dominant feeder. Ibis fed on individual pellets
one at a time, which could be observed by a distinctive backwards
motion of their head which propelled the pellet from the tip of its
bill to its throat.
Feeding sessions were excluded from the analysis if: 1)there were
major interruptions that resulted in truncated feeding sessions (e.g.
competition from other birds or disturbance by people); 2)other
bird species fed from the dish; or 3)dishes were upturned, despite
our precautions. Behavioural observations of ibis feeding at the
dish were within 5 m (birds were already habituated to human pres-
ence), typically around 10 m (following Murray and Shaw 2009).
After a feeding session, remaining pellets were collected and stored
in separate plastic bags before being weighed later that day.
Nutritional composition of naturalprey
To estimate the nutritional composition of the natural ibis diet, we
identiﬁed natural food items potentially consumed by ibis via lit-
erature search (Carrick 1959; Vestjens 1973; Barker and Vestjens
1990; Smith and Munro 2008; Chard 2015), and estimated their
nutritional compositions via the literature (Tacon et al. 1983; Jones
et al. 1996; Bernard et al. 1997; Dierenfeld et al. 2002; Finke
2002; Ahmed 2008; Xiaoming etal. 2010) and the USDA nutrient
database (US Department of Agriculture, Agricultural Research
Service, Nutrient Data Laboratory 2015) (Supplementary Table
S2). We excluded anthropogenic foods, either scavenged or deliber-
ately provisioned, from this analysis.
Mass of food consumed
We assessed the dierence in the mass of food consumed in grams
from each dish compartment using linear mixed models (LMMs)
implemented with the “lmer” function in the R package lme4
(Bates et al. 2014) following Machovsky-Capuska et al. (2016a).
All analyses were performed in R version 3.0.3 (R Core Team
2014). We set the response variable as the log-transformed (+0.5;
Yamamura 1999) mass in grams of food consumed after each feed-
ing trial. The predictor variable was a 3-level categorical variable
representing one of the three experimental foods oered (HP, HL,
and HC) per session. Signiﬁcance was inferred if 95% CIs did
not span zero. LMMs included a random factor for each feeding
session, to account for any between-session variance that may be
driven positional eects arising from how foods were presented.
Nutritional geometry of proportional
We used right-angled mixture triangles (RMT) to visualize the mac-
ronutrient ratio of foods selected by ibis groups at feeding stations
(full details of RMT methodology can be found in Raubenheimer
2011), where macronutrients were plotted on a percent metabo-
lizable energy basis and expressed as a proportion of total mac-
ronutrient-derived metabolizable energy. To do this, we applied
metabolizable energy conversion factors of 4 kcal/g for protein and
carbohydrate, and 9 kcal/g for lipid, to food compositions to esti-
mate metabolizable energy supplied by each macronutrient (Merrill
and Watt 1973). The metabolizable energy supplied by individual
macronutrients in each food was then plotted as a percentage of
the sum of total macronutrient energy. For simplicity, we express
the percent metabolizable energy of each macronutrient as: P (pro-
tein), L (lipid), and C (carbohydrate). HP pellets contained 67.6%
P, 25.4% L, and 7.0% C. HL pellets contained 2.4% P, 97.5% L,
and 0.1% C.HC pellets contained 13.5% P, 10.3% L, and 76.2%
C. Macronutrient composition estimates for natural foods were
plotted in the RMT using the same metabolizable energy conver-
sions to visualize the macronutrient space of their carnivorous diet.
Linear models of environmental factors affecting
proportional macronutrient selection
To determine whether rainfall and intraspeciﬁc competition
aected the proportions of P, L and C selected by ibis during feed-
ing sessions, we used linear models (LMs) implemented in the R
package compositions using an additive log-ratio (“alr”) approach
to analyzing closed compositions in a logistic geometry (“acomp”;
van den Boogaart and Tolosana-Delgado 2008; van den Boogaart
et al. 2014). The macronutrient C was treated as the denomina-
tor of the log-ratio, where two separate models were created with
either P or L in the numerator (“ln(P/C)” and “ln(L/C),” respec-
tively). We evaluated whether there was a relationship between the
total number of birds (“n_birds”) observed feeding from the food
dish during a feeding session and the proportion of macronutri-
ents selected. We included a quadratic term “n_birds^2” to evalu-
ate support for a non-linear relationship. We also explored whether
Representation of an experimental session showing: (a) a feeding station with an ibis feeding from one of three dish compartments each containing dierent
experimental pellets (HP, HL, or HC); and (b) a large group of ibis congregated around the feeding dish.
Page 3 of 9
total rain (mm) over the previous two days (“Rain2d”) inﬂuenced
macronutrient choice. Rainfall data was obtained from the Bureau
of Meteorology website (www.bom.gov.au) for the RBG weather
We performed a sensitivity analysis for Rain2d using data from
the nearby Observatory Hill station and obtained nearly identi-
cal results (Supplementary Table S3). We also evaluated whether
confounding factors, “Site” (a two-level categorical predictor; Hyde
Park vs. RBG), and “Middle_food” (three-level categorical predic-
tor denoting the contents of the middle compartment of the feed-
ing dish) aected the composition of foods consumed, but detected
no signiﬁcant eects (results not shown).
Mass of food consumed
On average, signiﬁcantly more HC pellets were consumed dur-
ing a feeding session (n = 61 sessions) than HP or HL pellets
(LMM est.HP-HC, CI= −0.912, −1.115 to −0.709; LMM est.HL-HC,
CI = −1.671, −1.874 to −1.468; Figure 2). Likewise, signiﬁcantly
more HP pellets were consumed than HL pellets (LMM est.HL-HP,
CI = −0.759, −0.962 to −0.556). Back-transformation indicated
that the average mass of HC (11.1 g) consumed per session was
almost 3 times that of HP (4.1g), and nearly 7 times greater than
HL (1.7g). The mean and mode number of birds feeding per ses-
sion was 5.3 (± 0.6 SE) and 4, respectively.
Analysis of proportional macronutrient
Right-angled mixture triangle
On a metabolizable energy basis, the ratio of macronutrients
selected by ibis during individual feeding trials (n=61) was usually
higher in C than P or L (Figure 3). Across all feeding events, the
arithmetic mean proportion of macronutrient energy selected dur-
ing feeding trials was 25% P (± 1.1 SE): 23% L (± 1.1): 52% C (±
1.8), which is higher in C and lower in L than would be expected
based on random feeding (26% P: 47% L: 27%C).
The macronutrient ratios of natural foods of ibis determined
from the literature were typical of those consumed by carnivores,
being moderate to high in P, low to moderate in L, and low or
devoid of C (Figure 3). Insects were a modest source of C, while
crustaceans and vertebrate prey contained little or negligible C.We
observed ibis to feed on earthworms several times (this was not
quantiﬁed), particularly on mornings following heavy rain, which
supports data from the literature (the eect of rain is statistically
explored in the next section); these prey are high in P and moderate
to low in L and C, although earthworms were among the natural
foods highest inC.
Linear models of factors affecting proportional
There was an increase in selection of P after rainfall, with ln(P/C)
showing a small but statistically signiﬁcant increase with Rain2d
(LM est.ln(P/C) = 0.006; Table 1; Figure 4). There was no signiﬁ-
cant eect of the number of birds foraging at the dish on ln(P/C).
Selection of L relative to C also increased slightly after rainfall, with
ln(L/C) showing a small but signiﬁcant relationship with Rain2d
(LM est.ln(L/C) = 0.005). There was also a signiﬁcant quadratic
relationship with the number of birds feeding and ln(L/C), where
ln(L/C) increased with n_birds (LM est.ln(L/C) = 0.182) followed by
a decrease with n_birds^2 (LM est.ln(L/C) = −0.008; Table 1; Figure
4). The quadratic relationship showed that ln(L/C) increased with
group size up to around 10 birds, after which it appeared to level
o (Figure 4).
Our study yielded several insights in to the nutritional ecology of
ibis in urban settings, demonstrating that ibis typically selected HC
pellets over HP and HL, and HP over HL. These results support
the hypothesis that ibis do not forage primarily for energy “per se”,
but distinguish between dierent macronutrient combinations. Ibis
selection from feeders was typically high in C, which is in contrast
to the typical macronutrient composition of their natural foods.
Competition for a locally concentrated food resource in the form of
our experimental food dish showed that ibis selected higher propor-
tions of L as group size increased, although only up to a point. Ibis
also displayed a behavioral response to rainfall, where they tended
to show an increased preference for P and L relative to C after peri-
The compositions of natural foods of ibis were relatively high in
P and L, and low or absent in C, which is consistent with a typical
carnivorous diet (Kohl etal. 2015). The diets of carnivores, how-
ever, have been considered to follow a gradient of omnivory based
on the increasing utilization of C (Remonti etal. 2015). For exam-
ple, on one end of the continuum omnivorous carnivores such as
grizzly bears (Ursus arctos) heavily consume C-rich foods (e.g. fruits)
when seasonally available (Coogan etal. 2014), while at the other
extreme obligate carnivores such as wolves and feral cats consume
very little C (~1–2%; Plantinga etal. 2011; Bosch etal. 2014). Our
analysis of the natural diet of ibis suggests that they typically con-
sume low to moderate levels of C (e.g. up to ~20% in earthworms)
in natural habitats. However, selection from our feeders indicated
that ibis have a high preference for C intake (76% C in HC pel-
lets). The carnivorous European badger (Meles meles; Remonti etal.
2011) showed a similar, although less extreme ﬂexibility in C intake,
consuming approximately a ﬁve-fold amount of C when in crop
lands depauperate in prey (Remonti etal. 2011).
Consumption (log grams+0.5)
HC HL HP
The log mean (+ 0.5) grams of food consumed by ibis per feeding trial
(n=61). Error bars are ± 1 SE.
Page 4 of 9
Coogan etal. • Macronutrient selection of urbanibis
Our study suggests that ibis display a high degree of food-com-
position generalism (“sensu” Machovsky-Capuska et al. 2016b),
being able to potentially feed on a range of natural foods, as well
as our experimental foods, which taken together vary widely in
macronutrient composition. In terms of the macronutrient com-
position on which ibis are able to subsist (i.e. their “fundamental
macronutrient niche”, “sensu” Machovsky-Capuska etal. 2016b),
ibis may be macronutrient generalists able to tolerate a wide range
of macronutrient intakes, or conversely they may be macronutri-
ent specialists that consume very dierent foods to arrive at a con-
sistent macronutrient intake (Machovsky-Capuska et al. 2016b).
For example, the pine marten (Martes martes) and European badger
consumed a wide variety and combination of foods between geo-
graphically distinct populations, yet each species maintained similar
macronutrient intake ratios across populations despite dietary dif-
ferences (Remonti etal. 2011, 2015). Such a situation is indicative
Mean feeder selection (+/– SE)
Random feeding (null hypothesis)
        
20 25 30 35 40 45 50 55
Protein (% energy)
Lipid (% energy)
[Carbohydrate (% energy)]
60 65 70 75 80 85 90 95
Right-angled mixture triangle (RMT) showing the macronutrient mixtures (green circles) and mixture space (green polygon) selected by ibis when oered
a dish containing three individually compartmentalized experimental foods. The grey-shaded polygon connecting the three experimental foods, HP (black
triangle), HF (black circle), and HC (hollow square), represents the experimental mixture space within which feeder meal compositions were constrained to
lie. Blue circles represent the macronutrient balance of a variety of “natural” foods of ibis identiﬁed from the literature. Earthworms are distinguished, as
they were documented to be an important food in our study area and were observed to be fed upon by ibis during ﬁeld work. Note that there are overlapping
data points for HC-only food selection which occurred during n=5 sessions (dark green circle). The arithmetic mean diet across all feeding trials is shown by
the black circle with bi-variate ±SE bars (for protein and lipid). The black square represents the macronutrient composition under the null hypothesis that ibis
selected the three foods in equal proportions or randomly.
Compositional linear model parameter estimates and SE predicting the log-ratio macronutrient proportions, ln(C/P) and ln(L/C),
selected by ibis during feeding sessions (n=61) as a quadratic function of the total number of birds observed eating food from the
experimental dish (n_birds + n_birds^2) and the total rain (mm) in the previous 2days (Rain2d)
Model parameters Estimate SE 95% CI lower 95% CI upper Adj. R2
(Intercept) −1.068 0.170 −1.408 −0.728
Rain2d 0.006 0.002 0.003 0.010
n_birds 0.067 0.055 −0.044 0.178
I(n_birds^2) −0.003 0.003 −0.010 0.004
(Intercept) −1.507 0.182 −1.872 −1.143
Rain2d 0.005 0.002 0.001 0.008
n_birds 0.182 0.059 0.063 0.301
I(n_birds^2) −0.008 0.004 −0.016 −0.001
Also given are 95% conﬁdence intervals (CI), and adjusted R-squared (Adj. R2) of model ﬁt. Non-intercept 95% CIs that do not include zero are shown in bold.
Page 5 of 9
of evolved homeostatic regulatory mechanisms for optimizing mac-
ronutrient intake dictating the patterns of foraging in these spe-
cies (Raubenheimer etal. 2016). A widespread investigation of the
nutritional preferences of ibis across geographically distinct popula-
tions (e.g. urban vs. rural) would yield insights into the fundamental
nutritional niche of the species.
A C-deprived diet reduced aggression and activity in Argentine
ants (Linepithema humile), suggesting that C-rich diets might enhance
competitive behaviour (Grover et al. 2007), and interestingly
Australian ecosystems may be rich in carbohydrates due to nutri-
ent poor soils (Low 2014). In our study, the non-linear relationship
between the number of birds feeding and the proportion of L,
relative to C, may suggest that food choice is inﬂuenced by group
competition in ibis. One potential mechanism behind this, is that
as group size increases feeding becomes more indiscriminate across
all individuals (scramble competition); however, it is more likely
in our study that more of the less preferred food (HL) was con-
sumed as the number of subordinate animals indiscriminately feed-
ing increased with group size (contest competition). The trend of
increasing L appeared to stabilize, however, when approximately
10 or more birds had fed from the dish, suggesting that the pro-
portion of L selected was converging towards the mean at larger
group sizes. This could indicate that the maximum number of
indiscriminate feeders able to feed at the dish had been reached
by this point, and that variance in lipid intake was being averaged
away as more birds fed from the dish. A future line of inquiry,
therefore, might be whether intraspeciﬁc competition for food
resources is a driver of dietary imbalance in animals that feed in
large groups such as ibis, and how such competition contributes to
evolutionarily adaptive responses to dietary imbalance (i.e. “rules of
compromise”; Raubenheimer and Simpson 1993). Theory predicts
that high-intensity intra-group competition for nutrients can result
in organisms altering their nutritional strategy (Senior etal. 2015).
Given the observed relationship between group size and macronu-
trient choice, ibis may be a good model system for future research
on the inﬂuence of conspeciﬁc competition on nutritional strategy.
Urban habitats can be highly unstable and do not completely
buer anthropogenic commensal species from seasonal variability
(Hulme-Beaman et al. 2016). Ibis appeared to select a wide range
of macronutrient ratios during feeding sessions following periods
of zero to little rainfall. After periods of rainfall, however, ibis dis-
played a slight preference for increased P and L consumption that
increased linearly with rainfall, which is more typical of the mac-
ronutrient composition of the natural foods (e.g. earthworms) that
are likely more abundant post-rainfall (Chard 2015). This is in
contrast to the prediction that ibis would select more C from our
050 100 150 200
Previous 2 day rainfall (mm)
050 100 150 200
Number of birds feeding
Results of compositional log-ratio based linear models of macronutrient proportions, ln(P/C) and ln(L/C), selected by ibis as a function of the predictor
variables: (a) the total rain (mm) in the previous 2days; and (b) the total number of birds observed feeding on experimental foods during a session. Blue lines
are ﬁtted values based on linear models, and the red lines show the mean ln(P/C) and ln(L/C) selected by ibis.
Page 6 of 9
Coogan etal. • Macronutrient selection of urbanibis
dish to redress a macronutrient imbalance after feeding on P- and
L-rich prey post-rainfall. One biological explanation is that ibis
are tuning their macronutrient preferences in keeping with envi-
ronmental resource availability post-rainfall to take advantage of
abundant resources. Such foraging behavior may be selected for in
highly-generalist feeders which encounter and opportunistically con-
sume a wide range of foods with diverse macronutrient composi-
tions, and which are able to tolerate a wide range of macronutrient
imbalance (Raubenheimer and Simpson 1999; Simpson etal. 2002;
Raubenheimer and Simpson 2003). Generalists with a wide diet
breadth have an elevated probability of subsequently encountering
foods with complementary nutritional properties, and taking advan-
tage of abundant yet nutritionally imbalanced resources may result
in higher ﬁtness. Alternatively, the physiological requirements of ibis
may shift to higher P and L following periods of rainfall for some
yet to be determined reason. Another possibility is that the observed
relationship is being driven by outliers in the data following the peri-
ods of extreme rainfall. More research is thus required to elucidate
the relationship between ibis macronutrient preferences and rainfall.
One explanation for the ibis’ preference for HC foods in experi-
mental sessions compared to their natural diet, may be in part due
to the relatively low proportion and availability of C in their natural
foods. Under natural conditions, the ibis appetite systems may have
evolved to preferentially consume C when present given its relative
rarity in the nutritional environment to which they are adapted. For
example, humans are believed to have adaptively evolved behav-
ioral and physiological regulatory systems that ﬁnd foods high in
C and L appetizing (Raubenheimer et al. 2014), because such
foods may have been rare in the ancestral environment (Speth and
Spielmann 1983; Cordain et al. 2000). In urban environments,
however, high-C food items are likely to be relatively more avail-
able than high-P and high-L foods (Coogan and Raubenheimer
2016). A species preference for high-C foods may therefore play a
role in human-wildlife conﬂict associated with anthropogenic food
resources. For instance, the grizzly bear is well known for incidences
of food-related conﬂict with humans and also has a preference for
high-C foods, especially when high-L foods are limiting (Coogan
and Raubenheimer 2016). An alternative explanation for our
results may thus be that ibis have an increased preference for C in
the urban environment, which may be due to a variety of factors
we did not have the opportunity to investigate (e.g. rapid local adap-
tation or phenotypic plasticity). The preference for C displayed by
relatively habituated urban ibis in our study may not be representa-
tive of the wild population. However, given the large population of
ibis in urban centres (8900 ibis in Sydney alone in 2008; Martin et
al. 2010), the preference for C may reﬂect the behaviour of a rela-
tively large proportion of the population. Ibis preference for high-C
therefore presents several lines of inquiry. Repeating studies such as
this across the ibis’ range and ecological contexts may yield insights
into the dynamics and evolution of nutritional preferences and the
factors involved in pre-adaptation and local-adaptation to urban
environments (McDonnell and Hahs 2015) and urban commensal-
ism (Hulme-Beaman et al. 2016).
CONCLUSIONS AND FUTURE RESEARCH
We have identiﬁed that urban ibis show a strong preference for C,
providing novel insights into the nutritional ecology of this suc-
cessful urban wildlife species. The inﬂuence of environmental (i.e.
across dierent contexts) and climate variables on the foraging
behavior of ibis will provide insights into how urban habitats have
inﬂuenced the nutritional preferences of this species, and more gen-
erally. Furthermore, we believe studies on urban nutritional ecology
can lead to a more holistic understanding of the general processes
underlying urban ecology and human-wildlife conﬂict, which can
in turn be applied in management strategies. A future priority is to
integrate data of free-ranging urban wildlife with controlled experi-
mental studies (e.g. in free-ranging enclosures). This combined
approach will allow us to better understand the nutritional dimen-
sions of foraging behaviour in complex and competitive nutritional
environments. We believe that ibis represent a potentially valuable
model system for the examining nutritional characteristics of a
highly successful urban wildlife generalist.
Supplementary data are available at Behavioral Ecology online.
S.C.P.C. was supported by the Natural Sciences and Engineering
Research Council (NSERC) of Canada, and an Australian
International Postgraduate Research Scholarship (IPRS) and
Australian Postgraduate Award (APA). GEMC is supported by the
Loxton research fellowship from the Faculty of Veterinary Science,
The University of Sydney.
We thank Tamara King and Richard Burman for assistance in the ﬁeld. We
thank Dr. Sonia Liu for assistance with experimental feeds.
Data accessibility: Analyses reported in this article can be reproduced using
data provided by Coogan etal. (2017).
Handling editor: Marc Thery
Ahmed MH. 2008. Evaluation of gambusia, Gambusia anis, ﬁsh meal in
practical diets for fry Nile tilapia, Oreochromis niloticus. J World Aquac Soc.
Aronson MF, La Sorte FA, Nilon CH, Katti M, Goddard MA, Lepczyk CA,
Warren PS, Williams NS, Cilliers S, Clarkson B, et al. 2014. A global
analysis of the impacts of urbanization on bird and plant diversity reveals
key anthropogenic drivers. Proc Biol Sci. 281:20133330.
Audet J-N, Ducatez S, Lefebvre L. 2016. The town bird and the country
bird: problem solving and immunocompetence vary with urbanization.
Behav Ecol. 27:637–644.
Barker RD, Vestjens WJM. 1990. The food of Australian birds: Volume 1,
non-passerines. Lyneham (Australia): CSIRO Australia.
Bates D, Mauchler M, Bolker B, Walker S. 2014. lme4: linear mixed-eects
models using Eigen and S4 [cited 2016 November]. Available from:
Beckmann JP, Berger J. 2003. Using black bears to test ideal-free distribu-
tion models experimentally. J Mammal. 84:594–606.
Bernard JB, Allen ME, Ullrey DE. 1997. Feeding captive insec-
tivorous animals: Nutritional aspects of insects as food.
Nutrition Advisory Group Handbook, Fact Sheet [cited 2015
November 10]. Available from: http://nagonline.net/801/
Bosch G, Hagen-Platinga EA, Hendriks WH. 2014. Dietary nutrient pro-
ﬁles of wild wolves: insights for optimal dog nutrition? Br J Nutr. 21:1–15.
Carrick R. 1959. The food and feeding habits of the straw-necked ibis,
Threskiornis spinicollis (Jameson), and the white ibis, T. Molucca (Cuvier), in
Australia. CSIRO Wildlife Research. 4:69–92.
Carrick R. 1962. Breeding, movements and conservation of ibises
(Threskiornithidae) in Australia. CSIRO Wildlife Research. 7:71–90.
Page 7 of 9
Chace JF, Walsh JJ. 2006. Urban eects on native avifauna: a review. Landsc
Urban Plan. 74:46–69.
Chard M. 2015. The mechanisms which drive the foraging decisions of
an urban exploiter [BSc (Hons) Thesis]. [Wollongong, NSW, Australia]:
University of Wollongong.
Coogan SC, Raubenheimer D, Stenhouse GB, Nielsen SE. 2014.
Macronutrient optimization and seasonal diet mixing in a large omni-
vore, the grizzly bear: a geometric analysis. PLoS One. 9:e97968.
Coogan SCP, Raubenheimer D. 2016. Might macronutrient requirements
inﬂuence grizzly bear-human conﬂict? Insights from nutritional geometry.
Coogan SCP, Machovsky-Capuska GE, Senior AM, Martin JM, Major
RE, Raubenheimer D. 2017. Data from: Macronutrient selection of free-
ranging urban Australian white ibis (Threskiornis moluccus). Dryad Digital
Cordain L, Miller JB, Eaton SB, Mann N. 2000. Macronutrient estimations
in hunter-gatherer diets. Am J Clin Nutr. 72:1589–1592.
Cowling SJ, Lowe KW. 1981. Studies of ibises in Victoria, I: records of
breeding since 1955. Emu. 81:33–39.
Dierenfeld ES, Alcorn HL, Jacobsen KL. 2002. Nutrient composition of
whole vertebrate prey (excluding ﬁsh) fed in zoos. Beltsville (MD): US
Department of Agriculture, Agricultural Research Service, National
Agricultural Library, Animal Welfare Information Center.
Epstein JH, McKee J, Shaw P, Micalizzi G, Daszak P, Kilpatrick AM,
Kaufman G. 2006. The Australian white ibis (Threskiornis molucca) as a res-
ervoir of zoonotic and livestock pathogens. Eco Health. 3:290–298.
Finke MD. 2002. Complete nutrient composition of commercially raised
invertebrates used as food for insectivores. Zoo Biol. 21:269–285.
Gilbert NI, Correia RA, Silva JP, Pacheco C, Catry I, Atkinson PW, Gill
JA, Franco AM. 2016. Are white storks addicted to junk food? Impacts
of landﬁll use on the movement and behaviour of resident white storks
(Ciconia ciconia) from a partially migratory population. Mov Ecol. 4:7.
Gordon IJ, Altwegg R, Evans DM, Ewen JG, Johnson JA, Pettorelli N,
Young JK. 2016. Reducing agricultural loss and food waste: how will
nature fare? Anim Conserv. 19:305–308.
Grover CD, Kay AD, Monson JA, Marsh TC, Holway DA. 2007. Linking
nutrition and behavioural dominance: carbohydrate scarcity limits
aggression and activity in Argentine ants. Proc Biol Sci. 274:2951–2957.
Hulme-Beaman A, Dobney K, Cucchi T, Searle JB. 2016. An Ecological
and Evolutionary Framework for Commensalism in Anthropogenic
Environments. Trends Ecol Evol. 31:633–645.
Jensen K, Mayntz D, Toft S, Clissold FJ, Hunt J, Raubenheimer D, Simpson
SJ. 2012. Optimal foraging for speciﬁc nutrients in predatory beetles.
Proc Biol Sci. 279:2212–2218.
Jones PL, De Silva SS, Mitchell BD. 1996. Eect of dietary protein content
on growth performance, feed utilization and carcass composition in the
Australian freshwater crayﬁsh, Cherax albidus Clark and Cherax destructor
Clark (Decapoda, Parastacidae). Aquacul Nutr. 2:141–150.
Kalnay E, Cai M. 2003. Impact of urbanization and land-use change on
climate. Nature. 423:528–531.
Kingsford RT, Johnson W. 1998. Impact of water diversions on colonially-
nesting waterbirds in the Macquarie Marshes of arid Australia. Colon
Kohl KD, Coogan SCP, Raubenheimer D. 2015. Do wild carnivores forage
for prey or nutrients? BioEssays. 37:701–709.
Low T. 2014. Where song began. Melbourne (Australia): Penguin.
Lowry H, Lill A, Wong BB. 2013. Behavioural responses of wildlife to
urban environments. Biol Rev Camb Philos Soc. 88:537–549.
Machovsky-Capuska GE, Senior AM, Zantis SP, Barna K, Cowieson AJ,
Pandya S, Pavard C, Shiels M, Raubenheimer D. 2016a. Dietary pro-
tein selection in a free-ranging urban population of common myna birds.
Behav Ecol. 27:219.227.
Machovsky-Capuska GE, Senior AM, Simpson SJ, Raubenheimer D. 2016b.
The Multidimensional Nutritional Niche. Trends Ecol Evol. 31:355–365.
Martin JM, French K, Major RE. 2007. The pest status of Australian white
ibis (Threskiornis molucca) in urban situations and the eectiveness of egg-
oil in reproductive control. Wild Res. 34:319–324.
Martin J, French K, Major R. 2010. Population and breeding trends of an
urban coloniser: the Australian white ibis. Wild Res. 37:230–239.
Martin JM, French K, Ross GA, Major RE. 2011. Foraging distances and
habitat preferences of a recent urban coloniser: the Australian white ibis.
Landscape Urban Plan. 102:65–75.
Martin J, French K, Major R. 2012. Behavioural adaptation of a bird from
transient wetland specialist to an urban resident. PLoS One. 7:e50006.
McDonnell MJ, Hahs AK. 2015. Adaptations and adaptedness of organ-
isms to urban environments. Annu Rev Ecol Evol Syst. 46:261–280.
McKinney ML. 2002. Urbanization, biodiversity, and conservation. Bio
Merrill AL, Watt BK. 1973. Energy value of foods: basis and derivation.
Washington, DC: Agriculture Handbook 74. Washington, DC: U.S.
Government Printing Oce.
Murray M, Cembrowski A, Latham ADM, Lukasik VM, Pruss S, St. Clair
CC. 2015. Greater consumption of protein-poor anthropogenic food by
urban relative to rural coyotes increases diet breadth and potential for
human-wildlife conﬂict. Ecography. 38:1235–1242.
Murray NJ, Shaw PP. 2009. Foraging behavior and success of Australian
white ibis (Threskiornis Molucca) in an urban environment. Notornis.
Nie Y, Zhang Z, Raubenheimer D, Elser JJ, Wei W, Wei F. 2014. Obligate
herbivory in an ancestrally carnivorous lineage: the giant panda and
bamboo from the perspective of nutritional geometry. Funct Ecol.
Plantinga EA, Bosch G, Hendriks WH. 2011. Estimation of the dietary
nutrient proﬁle of free-roaming feral cats: possible implications for nutri-
tion of domestic cats. Br J Nutr. 106 Suppl 1:S35–S48.
R Core Team. 2014. R: A language and environment for statistical comput-
ing [internet]. Vienna (Austria): R Foundation for Statistical Computing
[cited 2016 November]. Available from: http://www.R-project.org.
Raubenheimer D, Simpson SJ. 1993. The geometry of compensatory feed-
ing in the locust. Anim Behav. 45:953–964.
Raubenheimer D, Simpson SJ. 1999. Integrating nutrition: a geometrical
approach. Ent Exp Appl. 91:67–82.
Raubenheimer D, Simpson SJ. 2003. Nutrient balancing in grasshoppers:
behavioural and physiological correlates of dietary breadth. J Exp Biol.
Raubenheimer D. 2011. Toward a quantitative nutritional ecology: the
right-angled mixture triangle. Ecol Monographs. 81:407–427.
Raubenheimer D, Machovsky-Capuska GE, Gosby AK, Simpson S. 2015.
Nutritional ecology of obesity: from humans to companion animals. Br J
Nutr. 113 Suppl:S26–S39.
Raubenheimer D, Machovsky-Capuska GE, Chapman CA, Rothman JM.
2015. Geometry of nutrition in ﬁeld studies: an illustration using wild
primates. Oecologia. 177:223–234.
Raubenheimer D, Simpson SJ, Le Couteur DG, Solon-Biet SM, Coogan
SCP. 2016. Nutritional ecology and evolution of aging. Exp Gerontol.
Remonti L, Balestrieri A, Prigioni C. 2011. Percentage of protein, lipids,
and carbohydrates in the diet of badger (Meles meles) populations across
Europe. Ecol Res. 26:487–495.
Remonti L, Balestrieri A, Raubenheimer D, Saino N. 2015. Functional
implications of omnivory for dietary nutrient balance. Oikos.
Ross GA. 2004. Ibis in urban Sydney: a gift from Ra or a Pharaoh’s curse?
In: Lunney D, Burgin S, editors. Urban Wildlife: More Than Meets the
Eye. Mosman, NSW (Australia): Royal Zoological Society of New South
Wales. p. 148–152.
Secr. Conv. Biol. Div. 2012. Cities and biodiversity outlook. Montreal
(Canada): Secretariat of the Convention on Biological Diversity.
Senior AM, Charleston MA, Lihoreau M, Buhl J, Raubenheimer D,
Simpson SJ. 2015. Evolving nutritional strategies in the presence of
competition: a geometric agent-based model. PLoS Comput Biol.
Seto KC, Güneralp B, Hutyra LR. 2012. Global forecasts of urban expan-
sion to 2030 and direct impacts on biodiversity and carbon pools. Proc
Natl Acad Sci USA. 109:16083–16088.
Speth JD, Spielmann KA. 1983. Energy-source, protein metabolism, and
hunter-gatherer subsistence strategies. J Anthropol Archeol. 2:1–31.
Smith ACM. 2009. Population ecology of the Australian white ibis,
Threskiornis molucca, in the urban environment [PhD thesis]. Sydney, NSW
(Australia): University of Technology.
Smith ACM, Munro U. 2008. Cannibalism in the Australian pelican
(Pelecanus conspicillatus) and Australian white ibis (Threskiornis molucca).
Smith ACM, Munro U, Figueira WF. 2013. Modelling urban populations
of the Australian White Ibis (Threskiornis molucca) to infor m management.
Popul Ecol. 55:567–574.
Simpson SJ, Raubenheimer D. 2005. Obesity: the protein leverage hypoth-
esis. Obes Rev. 6:133–142.
Page 8 of 9
Coogan etal. • Macronutrient selection of urbanibis
Simpson SJ, Raubenheimer D. 2012. The nature of nutrition: a unifying
framework from animal adaptation to human obesity. Oxford: Oxford
Simpson SJ, Raubenheimer D, Behmer ST, Whitworth A, Wright GA.
2002. A comparison of nutritional regulation in solitarious- and gregari-
ous-phase nymphs of the desert locust Schistocerca gregaria. J Exp Biol.
Tacon AGJ, Staord EA, Edwards CA. 1983. A preliminary investigation
of the nutritive value of three terrestrial lumbricid worms for rainbow
trout. Aquaculture. 35:198–199.
van den Boogaart KG, Tolosana-Delgado R. 2008. “compositions”:
A uniﬁed R package to analyze compositional data. Comput Geosci.
van den Boogaart KG, Tolosana R, Matevz B. 2014. compositions:
Compositional Data Analysis. R package version 1.40-1 [cited
2016 November 07]. Available from: http://CRAN.R-project.org/
U. N.Dept. Econ. Soc. A. Pop. Div. 2015. World population prospects:
the 2015 revision, key ﬁndings and advance tables. Working Paper No.
ESA/P/WP.241. New York: United Nations.
US Department of Agriculture, Agricultural Research Service, Nutrient
Data Laboratory. 2015. USDA National Nutrient Database for Standard
Reference, Release 28. Version Current: September 2015 [cited 2015
November]. Available from: http://www.ars.usda.gov/nea/bhnrc/nd.
Vestjens WJM. 1973. Feeding of white ibis on freshwater mussels. Emu.
Xiaoming C, Ying F, Hong Z, Zhiyong C. 2010. Review of the nutritive
value of edible insects. In: Durst PB, Johnson DV, Leslie RN, Shono K,
editors. Proceedings of a workshop on Asia-Paciﬁc resources and their
potential for development. 2008 FEB 19–21; Chiang Mai, Thailand.
Bangkok (Thailand): Food and Agriculture Organization of the United
Nations Regional Oce for Asia and the Paciﬁc. p. 85–92.
Yamamura K. 1999. Transformation using (x + 0.5) to stabilize the variance
of populations. Res Popul Ecol. 41:229–234.
Page 9 of 9