PreprintPDF Available
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

Rapid urbanization is a major cause of habitat and biodiversity loss and human-animal conflict. While urbanization is inevitable, we need to develop a good understanding of the urban ecosystem and the urban-adapted species in order to ensure sustainable cities for our future. Scavengers play a major role in urban ecosystems, and often, urban adaptation involves a shift towards scavenging behaviour in wild animals. We carried out an experiment at different sites in the state of West Bengal, India, to identify the scavenging guild within urban habitats, in response to human provided food. Our study revealed a total of 17 different vertebrate species were identified across sites over 498 sessions of observations. We carried out network analysis to understand the dynamics of the system, and found that the free-ranging dog and common mynah were key species within the scavenging networks. This study revealed the complexity of scavenging networks within human-dominated habitats.
Content may be subject to copyright.
1
Scavengers in the human-dominated landscape an experimental study
Sourabh Biswas1, Tathagata Bhoumik2, Kalyan Ghosh3, Anamitra Roy1, Aesha Laheri1,
Sampita Sarkar1 and Anindita Bhadra1*.
Affiliation:
1 Department of Biological Sciences, Indian Institute of Science Education and Research Kolkata,
Nadia, West Bengal, India.
2 Department of Zoology, West Bengal State University
3 Department of Zoology, University of Burdwan
*Address for Correspondence
Behaviour and Ecology Lab, Department of Biological Sciences,
Indian Institute of Science Education and Research Kolkata
Mohanpur Campus, Mohanpur, Nadia
PIN 741246, West Bengal, India.
Phone no: +91 33 6136 0000 ext 1223
*Corresponding author
E-mail: abhadra@iiserkol.ac.in
Abstract
Rapid urbanization is a major cause of habitat and biodiversity loss and human-animal conflict.
While urbanization is inevitable, we need to develop a good understanding of the urban
2
ecosystem and the urban-adapted species in order to ensure sustainable cities for our future.
Scavengers play a major role in urban ecosystems, and often, urban adaptation involves a shift
towards scavenging behaviour in wild animals. We carried out an experiment at different sites
in the state of West Bengal, India, to identify the scavenging guild within urban habitats, in
response to human provided food. Our study revealed a total of 17 different vertebrate species
were identified across sites over 498 sessions of observations. We carried out network analysis
to understand the dynamics of the system, and found that the free-ranging dog and common
mynah were key species within the scavenging networks. This study revealed the complexity
of scavenging networks within human-dominated habitats.
Keywords:
Scavengers, Urban adaptation, Anthropogenic food source, Scavenging network, Key
species.
Introduction:
A recurrent and persistent theme of discussion across disciplines and in public fora today is
climate change and its impact on human lives as well as the ecosystem. While a wide spectrum
of research focuses on mitigation of climate change, disaster risk management and
conservation of natural ecosystems, relatively less research focuses on the urban ecosystem as
a habitat for biodiversity conservation. Rapid urbanization poses a threat to most wild species,
but some species are nevertheless able to cope with the changing environment and colonize the
new niches available in the urban ecosystem to thrive. While rapid urbanization is a very recent
phenomenon in the history of the earth, alterations to the environment by humans is perhaps as
old as human history. Anthropogenic food subsidies became available to animals around the
3
hunter-gatherer communities of early humans. In the Neolithic era, with the advent of
agriculture and domestication of wild animals, human settlements became a source of
additional nutrition for various commensal species (Agudoet et al., 2010; Chamberlin et al.,
2005). The explosion of the human population led to major alterations in habitats, through
deforestation, agricultural practices and production of a large amount of waste, which acted as
food subsidies for various species of animals (Oro et al., 2013).
Anthropogenic food subsidies, like refuse from food production, garbage dumps, fisheries
discards, crop leftovers, feeding restaurants for scavengers, feeding stations for animals, etc.
are predictable resources (Oro et al., 2013). Due to this predictable nature of the resources, they
have been exploited widely by a large number of urban-adapted species, which typically
demonstrate phenotypic and behavioural plasticity, variable niche adaptability and dispersal
strategies (Luna et al., 2021). This temporal and spatial predictability of the food subsidies
impact the increased survival, population breeding site availability, population growth and
body mass of urban scavengers (Oro et al., 2013). Scavengers are animals that feed on dead
and decaying matter. In the wild habitats, they typically feed on carcasses. However, in the
urban ecosystem, scavengers are generalist feeders that feed on any available food, typically
from anthropogenic sources. Waste generation in and around human settlements has increased
substantially in the last few decades, and the volume of waste is predicted to double within the
next twenty years (Hoornweg & Bhada-Tata 2012). Around the world, over 3 million tons of
waste are generated regularly, which is expected to reach the staggering amount of 6 million
tons per day by 2025 (Hoornweg et al., 2013). Needless to say, this is expected to impact the
population dynamics and behaviour of urban scavengers and the ecology of wild species
substantially.
4
A few studies have focused on the carrion-feeding (Corté S-Avizanda et al., 2012) species
associated with scavenging communities in habitats with different predictable anthropogenic
food subsidies (Oro et al., 2013). Some studies have focused on the birds associated with
landfills and large garbage dumps. Very little information is available about scavenging
communities in urban habitats in South Asia. Due to the nature of human lifestyle in the present
times, the urbanization gradient across habitats is no longer very steep, and clear demarcations
of urban and rural habitats no longer exist. Hence, it is better to consider landscapes in terms
of levels of anthropogenic usage, rather than a simple dichotomy of rural and urban, based on
lifestyles. Katlam & Prasad (2018) have investigated the scavengers in the forest fringe areas
in Uttarakhand associated with seasonal tourist influx. All of the studies have focused on
predictable food subsidies. However, scavenging response to unpredictable food subsidies in
human-dominated habitats is yet to be explored.
The structure of the community is an important factor for ecosystem functioning and an
essential driver for the stability of the community and its conservation (González et al., 2020).
Vertebrate scavengers in terrestrial habitats play a crucial role in ecosystem functioning as they
help accelerate decomposition and nutrient cycling, stabilize food webs, help to control disease
transmission and pest expansion (Beasley et al., 2019; Inger et al., 2016). Several studies have
described the community structure through network analysis (Chakraborty et al., 2021;
González et al., 2020; Sebastián-González et al., 2016) which is an effective tool to summarize
the community assemblage pattern numerically and compare communities statistically
(Bascompte & Jordano, 2007). Community structures are analyzed in nested patterns, using
specialization and centrality levels for understanding of the importance of different species in
the ecological community (Bascompte & Jordano, 2007; Sebastián González et al., 2020).
Understanding the structures of scavenging communities can be crucial in ecosystem
5
management and conservation, and can be an important factor contributing to the policy level
decisions to ensure sustainable urban management.
This study aimed to document the diversity of the scavenging guild associated with daily
human-generated waste by providing vegetarian and non-vegetarian food typical to Indian
households ad-hoc at various sites. This helped us to document the scavenging guild and also
study the response pattern of different species to the food provided. Finally, we analyzed the
network structure of scavengers in response to human generated food source and identified the
keystone scavenger species in the human-dominated habitat. Since free-ranging dogs (Canis
lupus familiaris) are an integral part of all human habitats in India, and have been adapted to
living among humans for perhaps the longest time, we tried to specifically understand the role
of the free-ranging dogs within this scavenging guild, and how they impact the diversity of
scavengers exploiting the ad-hoc resources created.
Methods
Study area
The experiment was conducted between 2020 and 2021 in different parts of West Bengal, India.
Experimenters choose sites randomly in and around the human habitation and away from
existing resources. The study sites were distributed in different parts of West Bengal, India
(Table S1, Figure S1).
The scavenging guild experiment
The specific sites for the experiment were selected away from the existing regular resource
sites that were seen to be used for scavenging by free-ranging dogs, through reconnaissance
6
surveys. The experimenter upturned a bowl of boiled rice or roti pieces mixed with either
vegetarian curry (VF) or non-vegetarian (egg, fish or chicken) curry (NVF). This is the most
common food of people across India, and home cooked food was used for all the experiments.
The bowls used were typically of a volume of approximately 400 ml. Video recording was
initiated before placing the food from a distance of 5-10 meters, and recording was continued
for one hour or until all the food provided was consumed by scavengers. The food type
(VF/NVF) for the first trial was picked randomly for each site, and once a food type was
chosen, the same was provided twice a day during morning and afternoon sessions for ten
consecutive days, followed by the other food type over the next ten days.
Data collection
The videos were decoded to extract the following information:
(i) Identity of scavengers: Any individual that was seen to approach and sniff and/or
eat the food was designated as a scavenger. The species identities of the scavengers
were noted.
(ii) Latency: The time taken for an individual to approach and interact with the food,
from the time of dropping the food, was considered as latency (in seconds).
Data on temperature, wind speed, cloud cover and rain corresponding to the experimental
sessions were collected from www.worldweather.com.
Statistical analysis
(i) Non-parametric tests: R statistical (R Studio) software was used for all the statistical
analyses. Shannon diversity index (H) of the scavenger community in each session
was calculated using the vegan package (Oksanen et al., 2007). Mann-Whitney U
7
test was used to compare scavenger diversity and latency of the first responders
(FR) across the sessions (Morning/Afternoon) and food type (VF/NVF). Diversity
and latency of the FR were compared across sites and scavengers as FR using the
Kruskal-Wallis test. Post-hoc Dunn's test was performed to check for any
significant difference in both latencies of FR and session diversity across sites and
scavengers.
(ii) Modeling: A GLMM (Generalized linear mixed model) analysis was conducted to
understand the effect of the first responder (species identity) and temperature on the
latency of the FR using normal distribution after log-transforming the data for
latency of the first responder using “lme4 (Bates et al., 2015) package of R.
Session diversity of the species was separated into three categories; low (0 < 0.5
H), medium (0.5 <1 H), high (1 H). Ordinal mixed logistic regression (OMLR)
was carried out to understand (a) the effect of the first responder species, sessions,
and temperature on the session diversity category using “ordinal” package
(Christensen RHB, 2019); (b) the effect of dogs as responders and sessions on the
session diversity category and (c) the effect of common mynas as FR and session
on the session diversity category. Experimental sites were considered as the random
effect. The null and full models were compared for all the models. Dispersion of
the models and residual diagnostics were checked using the “performance” package
of R (Lüdecke et al., 2021). The alpha level was kept at 0.05 throughout the
analysis.
(iii) Network analysis: Sites and scavenger interaction networks were built using the
“bipartite” package in R (Dormann et al., 2008). Scavengers and sites were placed
in columns and rows, respectively, to construct a quantitative interaction matrix.
Nestedness and network specialization were computed for the composite network
8
(combined VF and NVF) and also separately for VF and NVF. A weighted matrix
of scavenger response was used to calculate the ‘nestedness metric based on the
overlap and decreasing fill (NODF) (Almeida-Neto et al., 2008). NODF was
computed using ‘nestednodf’ command in the ‘vegan’ package (Oksanen et al.,
2007) in R. NODF values range between ‘0’ (no nestedness) to 100 (complete
nestedness). Complementarity specialization (H2) was computed for weighted
matrix (scavenger response) using the ‘H2fun’ command in the ‘bipartite’ package
in R. The range of (H2) values fall between ‘0’ (no specialization) and ‘1’ (complete
specialization).
A virtual exclusion method was executed to determine the significance of different scavengers
and sites in the respective networks. Each scavenger species was eliminated from the data to
compute the NODF and H2 values of the network, and then included back before eliminating
the next species. Using this method, the significance of each scavenger species was evaluated
in terms of network nestedness (NODF) and network specialization (H2).
Three species level indices normalized degree (ND), closeness centrality (CC) and
betweenness centrality (BC) were computed for the weighted composite network and
individually for VF and NVF networks. The degree of a species was divided by the number of
sites and was used to normalize the three indices (González et al., 2010). A relationship
between normalized degree, betweenness centrality and closeness centrality for both the
scavengers and sites were calculated using ‘Pearson’ correlation in R.
Results
Scavenging Guild
9
The experiment was performed in 498 different sessions, out of which 250 were in the morning,
and 248 were in the afternoon. A total of 240 sessions were completed with VF and 258
sessions with NVF. A total of 4403 individuals were recorded from the experimental sites.
Species responses were significantly higher (2 = 404.78, df = 1, p 0.05) for VF (2869), as
compared to NVF (1534). A total of seventeen vertebrate species of scavengers were recorded
in the study, out of which ten were avian species and six were mammals (Fig. S2, Fig. S3a).
Fifteen species of scavengers were recorded in VF (Fig. S3b), while sixteen species were
recorded in NVF (Fig. S3c). There was considerable overlap of species observed in the
different sites and for food types. Overall, House crows (Corvus splendens), sparrows (Passer
domesticus) and common mynas (Acridotheres tristis) were the most common scavengers. The
most abundant species responding to VF and NVF were sparrows, house crows, common
mynas and house crows, dogs, common mynas respectively. Cows (Bos taurus) only responded
to VF while white-breasted waterhen (Amaurornis phoenicurus) and domestic ducks (Anas
platyrhynchos domesticus) only responded to NVF.
Scavenger diversity of Sessions
Species diversity of session varied across the sites (Kruskal-Wallis rank-sum test: 2 = 173.36,
df = 14, p = <0.001). Scavenger diversity was relatively higher at three sites of Bongaon area
and the lowest at two sites of Durgapur (Fig. S4a).
Wilcoxon rank sum test revealed a significantly higher diversity of species (W= 27222.00, p =
0.01) in case of VF, as compared to NVF (Fig. S4b). The diversity of scavengers in the morning
and afternoon sessions were comparable (W= 29888.00, p = 0.45).
When cat (Felis catus), common myna and jungle myna (Acridotheres fuscus) were the FR, a
significant increase in the expected value of diversity on the log odds scale was observed,
10
given that all the other variables in the model were held constant, compared to dogs as FR
(Fig. S4c, Table. S2). This suggests that the dogs play a significant role in the scavenging
community. We witnessed a decrease in the expected value of diversity on the log odds scale
in case of temperature but the session didn’t affect the expected value of diversity (Table.
S2).
Dogs as no responder (NR) and second responder (SR) showed a significant increase in the
expected value of diversity category on the log odds scale compared to dogs as FR (Fig. 1,
Table. S3). The expected value of diversity category on the log odds scale positively
increased in the afternoon session compared to morning (Table. S3). Common myna was the
most ubiquitous scavenger that could be seen in all habitats. When common myna was the
FR, we did not see any effect on the expected value of diversity on the log odds scale, when
all the other variables in the model were held constant (Table. S4).
Latency
Kruskal-Wallis test revealed a significant difference in latency of the FR across sites (2 =
209.35, df = 14, p < 0.001). Latency was highest at three sites of Bongaon and lowest at two
sites of Durgapur (Kruskal-Wallis rank-sum test: 2 = 221.60, df = 14, p < 0.001) (Fig. S5a).
Latency of the FR were comparable across the morning and afternoon sessions (Wilcoxon
rank-sum test, W = 31188, p = 0.9). Latency of FR were also comparable across the veg and
non veg food (Wilcoxon rank-sum test, W = 30302, p = 0.7) (Fig. S5b). While all species were
included in the estimation of diversity, greater coucal, cow, jungle babbler, pied myna, goat
and squirrel were not considered as FR in the comparison of latency using the GLMM and
OMLR analysis that follow, as they contributed to 1% of the total sample size, and were
considered as outliers. Latency of FR were significantly different across species (Kruskal-
11
Wallis rank-sum test: 2 = 173.36, df = 14, p < 0.001). Latency of dogs as FR was the lowest
(2.32 ± 6.73s), whereas red vented bulbul showed the highest latency (11.41 ± 15.60s) (Fig.
S5c).
GLMM analysis revealed that the variation in latency of the FR was evident across
scavengers. Temperature positively affected the latency of the FR (Table. S5). The latency of
red-vented bulbul, cat, common myna, crow and jungle myna as FR were significantly higher
compared to the latency of dog as FR (Table. S5)
Female dogs exploited the resource more than male dogs (2 = 37.453, df = 1, p ≤ 0.0001). The
maximum number of species exploiting the resource was reached within 15 minutes of
providing the food, over a period of five days (Fig. 2).
The observed site and scavenger composite network (Fig. 3) combining VF and NVF showed
moderate nestedness (NODF= 55.84) and moderate specialization (H2 = 0.53). The network
structure did not differ for network nestedness (NODF = 52.54 and NODF = 56.02 in VF (Fig.
S6) and NVF (Fig. S7), respectively) and network specialization (H2 = 0.56 and H2 = 0.48,
respectively). H2 increased and NODF decreased after the virtual exclusion of dog and
common myna from the network (Fig. 3). H2 decreased after excluding house crow and
sparrow as scavenger species, as compared to the composite network (Fig. 4). Virtual exclusion
of cat, red vented bulbul, palm squirrel, goat, house crow, jungle crow, greater coucal and cow
increased NODF. We found a similar trend in the virtual exclusion of species in both VF and
NVF (see Fig. S8 for details).
12
70% of the scavenger species acted as a connector (BC > 0) in the combined network (Fig.
S9b), whereas for the VF and NVF networks these values were 67% and 63% respectively
(Fig. S9e & Fig. S9h) . This suggests that some connector species were unique to the two food
type networks, thus contributing to the composite higher representation of connectors in the
combined network. Dog and common myna share around 50% and 11% of the BC in the
composite network; 46% and 30% of the BC in VF; 27% and 27% BC in the NVF networks,
respectively (Fig. S9e & Fig. S9h). CC value of dog found to be the highest (0.068) among
the other scavengers in the composite network (Fig. S9c). Dog and common myna shared the
highest CC value (0.083 and 0.075) in both VF and NVF networks respectively (Fig. S9f &
Fig. S9i). The relationship between ND & CC and ND & BC for the scavenger species was
represented by a linear relationship (R = 0.59, p = 0.01 and R= 0.81, p < 0.01 respectively)
(Fig.5).
Discussion
The Anthropocene has been a period of large scale and far-reaching changes to the
environment, some of which have led to the present-day climate crisis. We need to take
cognizance of the fact that humans are constantly and irrevocably affecting the environment,
leading to habitat loss, biodiversity loss and behavioural changes in many species of the wild,
among others. The UN Sustainable Development Goals 2030, designed to achieve a sustainable
future for our planet, cannot be realized without the combined attention of scientists,
policymakers and members of the public. In the face of rapid urbanization, we need to work
towards a sustainable solution that would enable us to retain and nurture biodiversity in the
urban landscape. Understanding the ecology and behaviour of urban-adapted species can be
way forward towards this goal.
13
In this study, we documented the scavenging community in human-dominated habitats, which
were selected randomly, and had various levels of urbanization. While some sites were in areas
of high human flux like markets, other were within residential neighbourhoods. The locations
ranged from congested cities to relatively less congested townships. We documented a
reasonably large number of species exploiting the newly created unpredictable resource,
namely, human provided food. There was significant variation in the diversity and species
composition arising due to the nature of the food provided vegetarian or non-vegetarian,
while the basic component of the food was carbohydrates. It was interesting to see that
vegetarian food attracted more species than non-vegetarian food, which can most likely be
attributed to a higher number of avian species exploiting the VF. House crow, common myna,
sparrow and free-ranging dog were the most ubiquitous species, exploiting both vegetarian and
non-vegetarian food, but of these, only dogs were present across all the sites. These species
have been reported as scavengers from different parts of the world, and are also known for their
adaptation to human-dominated habitats (Noreen & Sultan, 2021).
Free-ranging dogs had a significant impact on the scavenger diversity of the sessions. When
dogs were the first to find the food source, the scavenger diversity of the session was lower
compared to the sessions in which dogs were not the first responders. This is probably because
dogs consumed most of the available food quickly, which did not leave much to be utilized by
other species. Moreover, virtual exclusion of dogs resulted in an increase of the NODF value
of the networks, underlining the key role that the dogs play in the scavenging network for both
VF and NVF. Common mynah was found to be the next most important scavenger species,
though they have not been reported in this role extensively yet. Mynahs are omnivorous birds,
and the current result is suggestive of urban adaptation in these birds through opportunistic
utilization of human generated resources.
14
Some species that were documented as scavengers in this study for the first time are jungle
myna (Acridotheres fuscus), pied myna (Gracupica contra), spotted dove (Spilopelia
chinensis), domestic goat (Capra aegagrus hircus), northern palm squirrel (Funambulus
pennantii), greater coucal (Centropus sinensis) and duck (Anas platyrhynchos domesticus).
Among these, the domestic goat and duck are accustomed to human food, due to their close
association with humans as domesticates, but species like the spotted dove and greater coucal
are neither domesticated, nor known to be scavengers. This observation thus raises questions
regarding the impact of human food waste on the dietary habits of these birds, and needs further
investigation.
Crow, common myna and sparrow comprised nearly 82% of the scavenging guild documented
responding to the VF. These species are communal foragers, and both inter and intra-species
co-feeding was observed during our experiments, which was probably the primary cause of
their success in monopolizing the food provided.
Scavenging from human-generated waste made these species successful in the human-
dominated landscape. Dogs are obligate scavengers due to their history of domestication
(Axelsson et al., 2013). They depend mostly on human-generated waste for survival (Bhadra
et al., 2016).
They are highly adapted to scavenging in the human-dominated landscape because of their high
cognitive ability to find food and superior olfaction (Sarkar et al., 2019) and have become
highly efficient scavengers in the human-dominated landscape. This might be the reason for
discovering food sources better than other scavenger species across our study sites.
15
Three sites in the Bongaon area were more of a rural habitat, whereas two sites in the Durgapur
area were highly urbanized. These sites in Bongaon had high latency, perhaps due to a lower
density of scavengers, or due to higher fear of humans. This needs further investigation. The
two sites in Durgapur had the fastest discovery time, probably because of a higher density of
dogs and less fear of humans in these areas. A significant drop in the diversity of species in the
presence of dogs as the first responder indicates the species' efficiency to monopolize human
subsidized food. Cortez Avizanda et al., (2012) also reported a similar kind of result, where the
diversity of the scavengers was relatively high before the arrival of the dominant specialist
scavenger species Griffon Vulture (Gyps fulvus), at the food source.
The composite, VF and NVF networks revealed a moderate nestedness and moderate
specialization compared to the highly nested network of scavenger assemblage in carrions from
all over the world (Sebastián González et al., 2020; Selva & Fortuna, 2007). Like plant-
pollinator mutualistic networks, the site-scavenger network can also be treated as a mutualistic
association, in which a scavenger species exploits the site for a particular type of resource for
their survival. In this case, the sites are not directly benefitted like the plants, but the reduction
of wats at a site can be considered as an indirect advantage for the site. The presence of the
scavengers maintain the ecosystem by increasing connectivity and stability in the food webs
(Inger et al., 2016; Rooney et al., 2006; Wilson & Wolkovich, 2011), disbursement of nutrients
in the ecosystem (Helfield & Naiman, 2001; Hewson, 1995; Schlacher et al., 2015) and
providing direct sanitary benefit to humans by removing potentially dangerous reservoirs of
bacteria (Ortiz & Smith, 1994; Vass A, 2001) and zoonotic pathogens fatal for humans and
other animals (Monroe et al., 2015).
16
In the nested community, networks are reported to be asymmetric and cohesive in nature
(Bascompte & Jordano, 2007; Sebastián González et al., 2020). Nestedness reflects the high-
density interactions of core taxa. For example, in plant-animal mutualisms, generalist species
form a dense core, and specialists are linked to it. In case of the scavenging guild, the highly
urban-adapted species, or more efficient scavengers, are expected to be common across sites,
while species with more limited scavenging capacity should occur in a subset of the sites (Selva
& Fortuna, 2007; Sebastián González et al., 2020). The competition for resources is higher in
the more nested networks of scavenger assemblages (Sebastián González et al., 2020). Such a
scenario can occur in multiple cases: (i) in scarce resource availability (Selva & Fortuna 2007),
(ii) the occurrence of specialized scavengers (e.g., vultures, Sebastián-González et al., 2016),
(iii) presence of dominant scavengers (e.g., black bears, Allen et al., 2014) that can monopolize
the resources or (iv) where the resources are relatively constant (Selva & Fortuna 2007;
Sebastián González et al., 2020). An increase in competition and a decrease in biodiversity is
observed in the networks with low nestedness (Bastolla et al., 2009; Basu P et al., 2016).
Contribution to network nestedness can be assessed as the control capacity of a species in the
mutualistic network (Cagua et al., 2019), and thus, network nestedness can be used to point out
the key species that have an impact on the overall network integrity (Campbell et al., 2012;
Cagua et al., 2019). The network becomes more stable with the decrease in network
specialization, making the network more resilient (Bascompte & Jordano, 2007; Basu et al.,
2021). Virtual exclusion of dogs and common myna from all the three kinds of networks
(composite, VF and NVF networks) reduced the network's nestedness and increased network
specialization. This suggests that the dog and common myna play important roles in the
scavenging networks and removing these generalist species from the network might decrease
the competition in the networks and make the network more vulnerable (Basu et al.,, 2021).
17
Species with high CC values impact the other species of the network, and species with a
positive value of BC in the network are essential for the network's cohesiveness (Estrada, 2007;
Martín González et al., 2010). The centrality indices identified the keystone species in the
networks across our studied sites. The architecture and persistence are impacted by the species
with a more central location in the network over the species at the edge (Cagua et al., 2019).
Dogs and common myna contributed the maximum in centrality indices, thus establishing them
as the most important species in the site-scavenger assemblage network. With a high
percentage (70%, 67% and 63%) of connector species in the composite, VF and NVF networks
reveal those species' importance in maintaining the site-scavenger network's integrity. Dog,
common myna and house crow were found to be the essential species contributing to the
stability of the VF network in the studied sites. However, at the same time, house crow and
sparrow were considered specialist scavenger species in some of the selected studied sites in
the VF and NVF site-scavenger network. In the composite network, dogs and common myna
were found to be the generalist species in the site-scavenger network, contributing as the most
important keystone species in the network. In contrast, other scavengers were found to be
specialists.
In the composite network, normalized degree (ND) showed a strong linear correlation with
both closeness and betweenness centrality. In both real-world and random networks, the
various centrality measures are expected to be correlated (Li et al., 2015). However, a strong
linear relationship between both ND and CC as well as ND and BC are not always observed in
networks in nature (Martín González et al., 2010). All these are centrality measures, that
indicate a node’s prominence in a network, and has the potential to play an important role in
information transfer within the network. In our case, the more connected scavenger species are
likely to influence the species diversity of the session.
18
Individual differences in response to anthropogenic food are influenced by age, sex or
hierarchical ranking (Oro et al., 2013), which is perhaps reflected in our observation of female
bias in dogs exploiting the food sources. Our findings indicate that female dogs are more
responsive toward food. The alternative hypothesis can be that females need more energy than
males. During pregnancy, females have many energy requirements. The study was performed
during dogs' pre-mating (July-September) and post-mating (October-December) seasons.
Female dogs might be more responsive to conserving energy for the gestation period. However,
a third explanation could be a female biased population of dogs in the areas of study, which is
unlikely, as an earlier study based on random sightings has reported no such bias in the
population (Sen Majumder et al., 2014). We need further studies to arrive at a conclusion
regarding this observation.
This study was a first attempt to document the scavenging community dependent on human
generated and provided food in human-dominated landscapes. While we documented the
commonly identified scavengers and were able to understand their roles in the scavenging
network, we also identified several species to exploit the given food, which have never been
documented as scavengers. We believe that this study builds a case for further investigation
into the identification of scavenging species and their interactions within human-dominated
habitats. Moreover, such studies can help us to understand how species might be undergoing
dietary shifts under the pressure of urbanization, to adapt to the more urban landscape. Such
knowledge will not only help to delve into the shifts in behaviour leading to urban adaptation,
but will also shed light on the evolutionary question of how some species might have adapted
to the changing environment to live around humans during the Anthropocene, exploiting the
rich and novel resources available among human generated waste.
19
Acknowledgements:
The authors would like to thank Dr Rubina Mondal, Dr Satyaki Mazumder and Dr Udipta
Chakraborty for their valuable inputs regarding data analysis. The authors would also like to
acknowledge the Indian Institute of Science Education and Research Kolkata for providing
infrastructural support. SB would like to thank the University Grants Commission, India for
providing him doctoral fellowship.
References:
Agudo, R., Rico, C., Vilà, C., Hiraldo, F., & Donzar, J. A. (2010). The role of humans in the
diversification of a threatened island raptor. BMC Evolutionary Biology, 10(1).
https://doi.org/10.1186/1471-2148-10-384
Allen, M. L., Elbroch, L. M., Wilmers, C. C., & Wittmer, H. U. (2014). Trophic facilitation
or limitation? Comparative effects of pumas and black bears on the scavenger community.
PLoS ONE, 9(7). https://doi.org/10.1371/journal.pone.0102257
Axelsson, E., Ratnakumar, A., Arendt, M. L., Maqbool, K., Webster, M. T., Perloski, M.,
Liberg, O., Arnemo, J. M., Hedhammar, Å., & Lindblad-Toh, K. (2013). The genomic
signature of dog domestication reveals adaptation to a starch-rich diet. Nature, 495(7441),
360364. https://doi.org/10.1038/nature11837
Ballejo, F., Lambertucci, S. A., Trejo, A., & de Santis, L. J. M. (2018). Trophic niche overlap
among scavengers in Patagonia supports the condor-vulture competition hypothesis. In Bird
Conservation International (Vol. 28, Issue 3, pp. 390402). Cambridge University Press.
https://doi.org/10.1017/S0959270917000211
20
Bascompte, J., & Jordano, P. (2007). Plant-animal mutualistic networks: The architecture of
biodiversity. In Annual Review of Ecology, Evolution, and Systematics (Vol. 38, pp. 567
593). https://doi.org/10.1146/annurev.ecolsys.38.091206.095818
Basu, P., Parui, A. K., Chatterjee, S., Dutta, A., Chakraborty, P., Roberts, S., & Smith, B.
(2016). Scale dependent drivers of wild bee diversity in tropical heterogeneous agricultural
landscapes. Ecology and Evolution, 6(19), 69836992. https://doi.org/10.1002/ece3.2360
Bates D, Mächler M, Bolker B, Walker S (2015) Fitting linear mixedeffects models using
lme4. J Stat Softw 67:148
Beasley, J. C., Olson, Z. H., Selva, N., & DeVault, T. L. (2019). Ecological Functions of
Vertebrate Scavenging (pp. 125157). https://doi.org/10.1007/978-3-030-16501-7_6
Bhadra, A., Bhattacharjee, D., Paul, M., Singh, A., Gade, P. R., Shrestha, P., & Bhadra, A.
(2016). The meat of the matter: a rule of thumb for scavenging dogs? Ethology Ecology and
Evolution, 28(4), 427440. https://doi.org/10.1080/03949370.2015.1076526
Bhattacharjee, D., & Bhadra, A. (2020). Humans Dominate the Social Interaction Networks
of Urban Free-Ranging Dogs in India. Frontiers in Psychology, 11.
https://doi.org/10.3389/fpsyg.2020.02153
Bhattacharjee, D., & Bhadra, A. (2021). Response to short-lived human overcrowding by
free-ranging dogs. Behavioral Ecology and Sociobiology, 75(7).
https://doi.org/10.1007/s00265-021-03052-x
Cagua, E. F., Wootton, K. L., & Stouffer, D. B. (2019a). Keystoneness, centrality, and the
structural controllability of ecological networks. Journal of Ecology, 107(4), 17791790.
https://doi.org/10.1111/1365-2745.13147
Cagua, E. F., Wootton, K. L., & Stouffer, D. B. (2019b). Keystoneness, centrality, and the
structural controllability of ecological networks. Journal of Ecology, 107(4), 17791790.
https://doi.org/10.1111/1365-2745.13147
21
Calcagno, V., Mouquet, N., Jarne, P., & David, P. (2006). Coexistence in a metacommunity:
The competition-colonization trade-off is not dead. In Ecology Letters (Vol. 9, Issue 8, pp.
897907). https://doi.org/10.1111/j.1461-0248.2006.00930.x
Campbell, C., Yang, S., Shea, K., & Albert, R. (2012). Topology of plant-pollinator networks
that are vulnerable to collapse from species extinction. Physical Review E - Statistical,
Nonlinear, and Soft Matter Physics, 86(2). https://doi.org/10.1103/PhysRevE.86.021924
Chakraborty, P., Chatterjee, S., Smith, B. M., & Basu, P. (2021). Seasonal dynamics of plant
pollinator networks in agricultural landscapes: how important is connector species identity in
the network? Oecologia, 196(3), 825837. https://doi.org/10.1007/s00442-021-04975-y
Chamberlain, C. P., Waldbauer, J. R., Fox-Dobbs, K., Newsome, S. D., Koch, P. L., Smith,
D. R., Church, M. E., Chamberlain, S. D., Sorenson, K. J., & Risebrough, R. (2005).
Pleistocene to recent dietary shifts in California condors.
www.pnas.orgcgidoi10.1073pnas.0508529102
Christensen RHB (2019). “ordinal—Regression Models for Ordinal Data .” R package
version 2019.12-10. https://CRAN.R-project.org/package=ordinal.
Corté S-Avizanda, A., Jovani, R., Carrete, M., Dona´zar, J. A., & Dona´zar, D. (2012).
Resource unpredictability promotes species diversity and coexistence in an avian scavenger
guild: a field experiment. In Ecology (Vol. 93, Issue 12).
Dormann, C. F. (2011). How to be a specialist? Quantifying specialisation in pollination
networks. In Network Biology (Vol. 1, Issue 1). www.iaees.orgArticle
Dormann, C. F. (2022). Using bipartite to describe and plot two-mode networks in R.
https://github.com/jgalgarra/bipartgraph,
Estrada, E., 2007. Characterization of topological keystone species local, global and‘‘meso-
scale’’ centralities in food webs. Ecol. Complex. 4, 48–57.
22
Gruber, B., Fründ, J., & Dormann, C. F. (2008). Introducing the bipartite Package: Analysing
Ecological Networks Otter Damage Monitoring View project Forces of Extinction in Reptiles
View project Introducing the bipartite Package: Analysing Ecological Networks (Vol. 8,
Issue 2). https://www.researchgate.net/publication/228861770
Helfield, J. M., & Naiman, R. J. (2001). Effects of salmon-derived nitrogen on riparian forest
growth and implications for stream productivity. In Reports Ecology (Vol. 82, Issue 9).
Hewson, R. (1995). Use of salmonid carcasses by vertebrate scavengers. Journal of Zoology,
235, 5365
Hoornweg, Daniel; Bhada-Tata, Perinaz. 2012. What a Waste : A Global Review of Solid
Waste Management. Urban development series;knowledge papers no. 15. World Bank,
Washington, DC. © World Bank. https://openknowledge.worldbank.org/handle/10986/17388
License: CC BY 3.0 IGO.
Hoornweg, D., Bhada-Tata, P. & Kennedy, C. Environment: Waste production must peak this
century.Nature 502, 615617 (2013). https://doi.org/10.1038/502615a
Inger, R., Cox, D. T. C., Per, E., Norton, B. A., & Gaston, K. J. (2016). Ecological role of
vertebrate scavengers in urban ecosystems in the UK. Ecology and Evolution, 6(19), 7015
7023. https://doi.org/10.1002/ece3.2414
Jagiello, Z., Dylewski, Ł., Tobolka, M., & Aguirre, J. I. (2019). Life in a polluted world: A
global review of anthropogenic materials in bird nests. In Environmental Pollution (Vol. 251,
pp. 717722). Elsevier Ltd. https://doi.org/10.1016/j.envpol.2019.05.028
Katlam, G., Prasad, S., Aggarwal, M., & Kumar, R. (2018). Current Science Association
Trash on the menu. 115(12), 23222326. https://doi.org/10.2307/26978598
Li, C., Li, Q., van Mieghem, P., Stanley, H. E., & Wang, H. (2015). Correlation between
centrality metrics and their application to the opinion model. European Physical Journal B,
88(3), 113. https://doi.org/10.1140/epjb/e2015-50671-y
23
Lima, M., Harrington, R., Saldañ, S., Estay, S., Lima, M., Saldaña, S., & Estay, S. (n.d.).
Non-linear feedback processes and a latitudinal gradient in the climatic effects determine
green spruce aphid outbreaks in the UK. https://doi.org/10.1111/j.2008.0030-1299.16615.x
Luna, Á., Romero-Vidal, P., & Arrondo, E. (2021). Predation and scavenging in the city: A
review of spatio-temporal trends in research. In Diversity (Vol. 13, Issue 2, pp. 116). MDPI
AG. https://doi.org/10.3390/d13020046
Martín González, A. M., Dalsgaard, B., & Olesen, J. M. (2010). Centrality measures and the
importance of generalist species in pollination networks. Ecological Complexity, 7(1), 3643.
https://doi.org/10.1016/j.ecocom.2009.03.008
Monroe, B. P., Doty, J. B., Moses, C., Ibata, S., Reynolds, M., & Carroll, D. (2015).
Collection and utilization of animal carcasses associated with zoonotic disease in Tshuapa
district, the democratic Republic of the Congo, 2012. Journal of Wildlife Diseases, 51(3),
734738. https://doi.org/10.7589/2014-05-140
Nielsen, A., & Bascompte, J. (2007). Ecological networks, nestedness and sampling effort.
Journal of Ecology, 95(5), 11341141. https://doi.org/10.1111/j.1365-2745.2007.01271.x
Noreen, Z., & Sultan, K. (2021). Population explosion and behavioural changes of
opportunist wild avifauna at a landfill at gujranwala in northeastern punjab: A baseline
deviation study. Pakistan Journal of Zoology, 53(6), 22552267.
https://doi.org/10.17582/journal.pjz/20200211050231
Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, Suggests MASS
(2007) The Vegan package. Commun Ecol Package 10:631637
Olson, Z. H., Beasley, J. C., Devault, T. L., & Rhodes, O. E. (2012). Scavenger community
response to the removal of a dominant scavenger. Oikos, 121(1), 7784.
https://doi.org/10.1111/j.1600-0706.2011.19771.x
24
Oro, D., Genovart, M., Tavecchia, G., Fowler, M. S., & Martínez-Abraín, A. (2013).
Ecological and evolutionary implications of food subsidies from humans. Ecology Letters,
16(12), 15011514. https://doi.org/10.1111/ele.12187
Ortiz, N. E., & Smith, G. R. (1994). The production of Clostridium botulinum type A, B and
D toxin in rotting carcasses. Epidemiology and Infection, 113(2), 335343.
https://doi.org/10.1017/S0950268800051761
Rooney, N., McCann, K., Gellner, G., & Moore, J. C. (2006). Structural asymmetry and the
stability of diverse food webs. Nature, 442(7100), 265269.
https://doi.org/10.1038/nature04887
Sarkar, R., Sau, S., & Bhadra, A. (2019). Scavengers can be choosers: A study on food
preference in free-ranging dogs. Applied Animal Behaviour Science, 216, 3844.
https://doi.org/10.1016/j.applanim.2019.04.012
Schlacher, T. A., Weston, M. A., Schoeman, D. S., Olds, A. D., Huijbers, C. M., & Connolly,
R. M. (2015). Golden opportunities: A horizon scan to expand sandy beach ecology.
Estuarine, Coastal and Shelf Science, 157, 16. https://doi.org/10.1016/j.ecss.2015.02.002
Sebastián-González, E., Moleón, M., Gibert, J. P., Botella, F., Mateo-Tomás, P., Olea, P. P.,
Guimarães, P. R., & Sánchez-Zapata, J. A. (2016). Nested species-rich networks of
scavenging vertebrates support high levels of interspecifi c competition. In Ecology (Vol. 97,
Issue 1).
Sebastián-González, E., Morales-Reyes, Z., Botella, F., Naves-Alegre, L., Pérez-García, J.
M., Mateo-Tomás, P., Olea, P. P., Moleón, M., Barbosa, J. M., Hiraldo, F., Arrondo, E.,
Donázar, J. A., Cortés-Avizanda, A., Selva, N., Lambertucci, S. A., Bhattacharjee, A.,
Brewer, A. L., Abernethy, E. F., Turner, K. L., … Sánchez-Zapata, J. A. (2020). Network
structure of vertebrate scavenger assemblages at the global scale: drivers and ecosystem
functioning implications. Ecography, 43(8), 11431155. https://doi.org/10.1111/ecog.05083
25
Selva, N., & Fortuna, M. A. (2007). The nested structure of a scavenger community.
Proceedings of the Royal Society B: Biological Sciences, 274(1613), 11011108.
https://doi.org/10.1098/rspb.2006.0232
Sen Majumder, S., Chatterjee, A., & Bhadra, A. (2014). A dog’s day with humans – time
activity budget of free-ranging dogs in India (Vol. 106, Issue 6).
https://www.jstor.org/stable/24102275
Sengupta, A., McConkey, K. R., & Kwit, C. (2022). The influence of provisioning on
animal-mediated seed dispersal. Oikos, 2022(2). https://doi.org/10.1111/oik.08276
Sharma, N., Gadagkar, R., & Pinter-Wollman, N. (2022). A reproductive heir has a central
position in multilayer social networks of paper wasps. Animal Behaviour, 185, 2136.
https://doi.org/10.1016/j.anbehav.2021.12.011
Ulrich, W., Almeida-Neto, M., & Gotelli, N. J. (2009). A consumer’s guide to nestedness
analysis. In Oikos (Vol. 118, Issue 1, pp. 317). https://doi.org/10.1111/j.1600-
0706.2008.17053.x
Valente, T. W., Coronges, K., Lakon, C., & Costenbader, E. (n.d.). How Correlated Are
Network Centrality Measures?
Wilson, E. E., & Wolkovich, E. M. (2011). Scavenging: How carnivores and carrion structure
communities. In Trends in Ecology and Evolution (Vol. 26, Issue 3, pp. 129135).
https://doi.org/10.1016/j.tree.2010.12.011
Figures with legends:
26
Figure 1: A box and whiskers plot showing the effect of dog response order on the session
diversity. FR First responder, SR Second responder and TR Third responder, RR Rest
of the positions of Responder; NR No responder. The black line represents the median value,
the open diamond marks the mean, the rectangle shows the 25th and 75th quartile of the data
and the whiskers represent the data range.
27
Figure 2: Species accumulation curve. a) the number of unique species saturation over time
(minutes); b) the number of unique species saturation over the days of the experiment.
28
Figure 3:
The bipartite graph represents the composite (VF and NVF combined) network structure of
scavengers in response to human generated food source in different sites of West Bengal.
Each upper rectangle represents a scavenger species and lower rectangle represents a site.
The size of the rectangle reflects the no of times a species appears in the network as
responder. The lines connect each site with the scavenger species responding at that site, and
the thickness of the lines show the number of individuals responding at the respective sites.
Figure 4.
This figure shows the results of virtual exclusion of scavengers describing the contribution of
different scavenger species in the network structure: a) network specialization (H2 ) and b)
network nestedness (NODF) H2 and NODF values of the respective networks without any
29
exclusion and the dots represent the corresponding values for exclusion of each scavenger
species.
Figure 5. a) The relationship between normalized degree (ND) and betweenness centrality
(BC) of scavenger species; b) The relationship between normalized degree (ND) and closeness
centrality (CC) of scavenger species.
Supplementary Information
Table S1: List of experimental sites. The table shows the various sites (with GPS locations)
in which sampling was carried within the state of West Bengal, India, and the type of habitat
(urban/suburban/rural) that they represented.
30
Place
Site
GPS coordinate
Bakkhali
Panchmathani beach
21.560129, 88.262594
Bakkhali beach
21.560129, 88.267003
Amaravati
21.560558, 88.266499
Gaighata
Jaleswar
22.928689, 88.709274
Srimantapur
22.931022, 88.722046
Narega
Das para spot 1
23.63486 87.96383
Das para spot 2
23.63443, 87.96361
Bangaon
Puraba para
23.051473, 88.845214
Nichu Para
23.052029, 88.842575
Math para
23.049635, 88.842064
IISER-Kolkata
ICVS
22.964909,88.3862325
Behala
Bakultala
22.485348, 88.301340
Durgapur
Shyampur site 1
23.2832046, 87.191320
Shyampur site 2
23.283362, 87.1910723
Shyamnagar
Shibpur
22.8353262,88.3862325
31
Figure S1: Map of study sites. The figure shows the various sites in which sampling was
carried within the state of West Bengal, India.
32
Figure S2: Scavengers responded in study sites:
This figure shows the no of scavengers found across different study sites.
a) House sparrow (Passer domesticus). b) Spotted dove (Spilopelia chinensis). c) House crow
(Corvus splendens). d) Common myna (Acridotheres tristis). e) Jungle crow (Corvus
culminates). f) Jungle babbler (Turdoides striatus). g) Jungle myna (Acridotheres fuscus). h)
White-breasted waterhen (Amaurornis phoenicurus). i) Red-vented bulbul (Pycnonotus
cafer). j) Greater coucal (Centropus sinensis). k) Pied starling (Gracupica contra). l)
Domestic duck (Anas platyrhynchos domesticus). m) Goat (Capra aegagrus hircus). n) Cat
(Felis catus). o) Cow (Bos taurus). p) Palm squirrel (Funambulus pennantii).
q) Dog (Canis lupus familiaris).
ab c d
efg h
ijklm
no p q
©Sourabh Biswas
© Kushankur Bhattacharyya
© Kushankur Bhattacharyya
© Kushankur Bhattacharyya
© Kushankur Bhattacharyya © Afsar Nayakkan © Mohammad Mahdhi Karim © Hobbi fotowiki
© J. M. Garg
© J. M. Garg © J. M. Garg
© Charles J. Sharp © Aswin06k © Asive Chowdhuory
© Bocken Inagory © Mammal watcher © Nidhi pious996
33
Figure S3: Response pattern of scavengers in study sites. This figure shows the no of
scavengers found across different study sites. a) Depicts the scavengers found in both VF and
VNF combined in the study sites; b) shows the scavengers found only in VF and c) shows the
scavengers responded in NVF. The color intensity (black-grey) of the heatmap, figures (a-c)
represent the dominance of species response high to low.
34
Figure S4: Difference in session diversity (H). a) Describes the difference of session
diversity across sites; b) shows the difference of session diversity between veg (VF) and non
veg (NVF) food type. Whereas, c) shows the impact of the first responder species on diversity
of the session.
Table S2: Effect of first responder species (scavengers) on the session diversity.
Ordinal mixed logistic regression (OMLR) analysis.
Model (Session diversity category (High/Medium and Low) ~ Scavengers as first responder
35
+ temperature + Session (Morning/ Afternoon), random factor = Study sites)
Location coefficients:
Fixed effects
Estimate
z value
Pr(>|z|)
Bulbul
0.80
0.9644
0.3349
cat
1.41
2.6506
0.0080
Common myna
0.69
2.3311
0.0197
House crow
0.79
2.1111
0.0348
Jungle myna
1.60
2.4203
0.0155
sparrow
0.42
0.6392
0.5227
temperature
-0.11
-2.2031
0.0276
noon (session)
0.32
1.6002
0.1096
Threshold coefficients:
Estimate
z value
High | Low
-1.9987
-9.8496
Low | Medium
1.6206
8.7513
Random effects
Group
Varience
Standard deviation
Place
0.2278142
0.4772988
log-likelihood
-402.5226
AIC
821.0452
Condition number of Hessian
149.2332
Table S3: Effect of dog response order on session diversity.
Ordinal mixed logistic regression (OMLR) analysis.
Model (Session diversity category (High/Medium and Low) ~ Dog response order (First
responder/ Second responder/ Third responder/ Rest of the responder and No responder)
+ Session (Morning/ Afternoon), random factor = Study sites)
Location coefficients:
Fixed effects
Estimate
Standard error
z value
Pr(>|z|)
Noon (session)
0.4427
0.1916
2.3107
0.020851
36
No Responder (NR)
1.0825
0.2267
4.7750
1.80E-06
Rest of Responder (RR)
-0.0197
0.4951
0.0397
0.968315
Second Responder (SR)
1.3127
0.6122
2.1441
0.032024
Third Responder (TR)
-0.4048
1.1842
-0.3419
0.732463
Threshold coefficients:
Estimate
Standard error
z value
high|low
-1.9987
0.2029
-9.8496
low|medium
1.6206
0.1852
8.7513
Random effects
Group
Varience
Standard deviation
place
0.06402972
0.2530409
log-likelihood
-402.5226
AIC
821.0452
Condition number of Hessian
149.2332
Table S4: Effect of Common myna response order on session diversity.
Ordinal mixed logistic regression (OMLR) analysis.
Model (Session diversity category (High/Medium and Low) ~ Common myna response order
(First responder/ Second responder/ Third responder/ Rest of the responder and No responder)
+ Session (Morning/ Afternoon), random factor = Study sites)
Location coefficients:
37
Fixed effects
Estimate
Standard error
z value
Pr(>|z|)
Noon (session)
0.4971
0.1998
2.4875
0.0128632
No Responder (NR)
-0.5570
0.3616
-1.5406
0.1234131
Rest of Responder (RR)
0.3089
0.4689
0.6586
0.5101218
Second Responder (SR)
0.7410
0.4900
1.5123
0.1304526
Third Responder (TR)
0.5647
0.6622
0.8526
0.3938569
Threshold coefficients:
Estimate
Standard error
z value
high|low
-2.6135
0.3773
-6.9262
low|medium
1.0006
0.3451
2.8994
Random effects
Group
Varience
Standard deviation
place
0.3313624
0.5756408
log-likelihood
-389.2415
AIC
794.4831
Condition number of Hessian
42.75265
38
Figure S5: Difference in latency of first responder (FR).
Figure a) describes the difference of latency of FR across experimental sites. This Box and
Whisker plot also describe the food discovery time of scavenger species in study sites. Figure
b) shows the difference of latency of FR between veg (VF) and non veg (NVF) food type.
Whereas, c) describes the difference of latency of FR scavengers.
Table S5: Effect of scavenger species on the food discovery time (Latency of FR)
Table showing the outcomes of GLMM analysis:
39
Model log( Latency of FR) ~ Scavenger species + Temperature + (1| study sites)
Fixed effects
Estimate
Standard error
df
t value
Pr(>|t|)
(Intercept)
-1.86486
0.57747
415.84
-3.229
0.00134 **
Bulbul
0.51253
0.26279
464.21
1.950
0.05173 .
Cat
1.03221
0.18912
464.54
5.458
7.85e-08 ***
Common myna
0.65557
0.10672
468.82
6.143
1.73e-09 ***
House crow
0.54489
0.13824
471.98
3.941
9.32e-05 ***
Jungle myna
1.19308
0.22937
463.94
5.202
2.97e-07 ***
Sparrow
0.36998
0.25370
473.00
1.458
0.14541
Temperature
0.04138
0.01785
467.92
2.318
0.02089 *
Random effects
Groups
Name
Variance
Standard deviation
place
Intercept
0.3897
0.6243
Residual
0.5374
0.7331
AIC
1142.959
BIC
1184.718
*** P < 0.0001, * P < 0.01, . P < 0.05, P < 0.1
40
Figure 6:
The bipartite graph represents the VF network structure of scavengers in response to human
generated food source in different sites of West Bengal. Each upper rectangle represents a
scavenger species and lower rectangle represents a site. The size of the rectangle reflects the
no of times a species appears in the network as responder. The line match scavenger species
responding a specific site, and the width of the line shows the number of individuals
responding that specific site.
 
 
 
  
 
 


   
    
 
 
 
 





 

      
41
Figure S7:
The bipartite graph represents NVF network structure of scavengers in response to human
generated food source in different sites of West Bengal. Each upper rectangle represents a
scavenger species and lower rectangle represents a site. The size of the rectangle reflects the
no of times a species appears in the network as responder. The line match scavenger species
responding a specific site, and the width of the line shows the number of individuals
responding that specific site.
  

 
   



 
  
 

 
 
     




 
 
 
   
 
42
Figure S8: Virtual exclusion of species.
This figure shows the results of virtual exclusion scavengers describing the contribution of
different scavenger species in the network structure. Fig (a & b) depict the network nestedness
(NODF) and network specialization (H2 ) in veg food whereas figure (c & d) shows the the
network nestedness (NODF) and network specialization (H2 ) in non veg food. The dashed
line represents the NODF and H2 value of composite network without any exclusion.
43
Figure S9: Species level indices value of scavengers.
This figure shows the Normality degree (ND), Betweenness centrality (BC) and Closeness
centrality (CC) of different scavenger species. Figure (a-c) represent the composite network
(CMP), figure (d-f) represent the veg food network (VF) and figure (g-i) represent the non veg
food network (NVF) ND, BC and CC accordingly.
44
Article
Full-text available
Acquiring knowledge about the environment is crucial for survival. Animals, often driven by their exploratory tendencies, gather valuable information regarding food resources, shelter, mating partners, etc. However, neophobia, or avoiding novel environmental stimuli, can constrain their exploratory behaviour. While neophobia can reduce potential predation risks, decreased exploratory behaviour resulting from it may limit the ability to discover highly rewarding resources. Dogs (Canis familiaris) living in semi-urban and urban environments as free-ranging populations, although subject to various selection forces, typically have negligible predation pressure. These dogs are scavengers in human-dominated environments; thus, selection against object-neophobia can provide benefits when searching for novel food resources. Although captive pack-living dogs are known to be less neophobic than their closest living ancestors, wolves (Canis lupus), little is known about free-ranging dogs' behavioural responses to novel objects, particularly in foraging contexts. Using an object choice experiment, we tested 259 free-ranging dogs from two age classes, adult and juvenile, to investigate their object-neophobia in a scavenging context. We employed a between-subject study design, providing dogs with a familiar and a potentially novel object, both baited with equal, hidden food items. Adult and juvenile dogs significantly inspected the novel object first compared to the familiar one, even when the hidden food item was partially visible. To validate these findings, we compared novel objects with different strengths of olfactory cues (baited vs. false-baited) and found that they were inspected comparably by adults and juveniles. No significant differences were found in the latencies to inspect the objects, suggesting that free-ranging dogs may still be cautious when exploring their environments. These results indicate that free-ranging dogs, evidently from an early ontogenetic phase, do not show object-neophobia, as demonstrated by their preference for novel over familiar food sources. We conclude that little to no constraint of neophobia on exploratory behaviour in semi-urban and urban-dwelling animals can guide foraging decision-making processes, providing adaptive benefits.
Article
Full-text available
Interaction with its immediate environment determines the ecology of an organism. Short-lived perturbations in the habitat can adversely affect both wild and domesticated species. When such disturbances are unpredictable, they are more challenging to tackle. Therefore, specific strategies become essential for species to overcome adversities that can alter their dynamics. Species living in the proximity of humans experience significant anthropogenic disturbances. Response to such a disturbance, facilitated by human overcrowding, was studied in an urban-adapted species. Free-ranging dogs (Canis familiaris) found in developing countries can inhabit both urban and rural areas. Dogs present in urban habitats experience tremendous anthropogenic pressure. It is known that human flux can predict variability in their behavior in different ‘microhabitats’. We investigated the effect of a sudden massive surge in human footfall, during a major festival, on their abundance, distribution and behavior. While both abundance and general activity levels declined, vigilance behavior heightened during the festival. The behavioral activity was restored post-event, but dog abundance remained significantly reduced compared to pre-festival levels. Furthermore, we found variations in dog distribution across the sessions with higher dispersion during the festival. Conclusively, this study provides substantial quantitative evidence on the extent to which humans can influence even the behavior of animals well-adapted to urban environments. Significance statement Changes in the immediate environment drive organisms to build efficient adaptive strategies. These changes could be both natural, like flood and storm, or anthropogenic, like human overcrowding. The latter especially bestows immense stress on urban living animals, i.e., those that live in close proximity to human settlements. Using an observational study, we show that a sudden surge in human footfall negatively impacts a highly urban-adapted animal, the free-ranging dogs. Dogs lower behavioral activity, become more vigilant and modify their spatial distribution lowering their abundance in such areas. Though short-lived in nature, these disturbances can leave a long-term residual impact on their population. In conclusion, sudden human-induced perturbations can impact even those animals that are considered highly adapted to urban conditions. They are seen to adopt varied strategies to withstand such disturbances in their immediate environment, thus altering their behavioral dynamics.
Article
Full-text available
Farm habitat enrichment is crucial for sustainable production of pollinator-dependent crops. Correct choice of crop and non-crop plant species in habitat management support resilient pollinator fauna and effective pollination service delivery. We identify key network metrices to recognize suitable crop and non-crop plant species for farm habitat enrichment. We also highlight the importance of seasonal variation of the key plant and pollinator species that will crucially inform farm management. Crop species played a key role in maintaining plant–pollinator network integrity. In contrast to the conventional practice of focussing on non-crop plants for pollination service restoration, we find crop plants across seasons hold a key role in maintaining healthy plant–pollinator networks. Our study highlights the importance of non-bee pollinators especially, flies and butterflies in sustaining healthy plant–pollinator network. Bees were important as connector species and controlled other species in the network. Only 16.67% bees and 33.33% of plant species acted as connector species. Our study also shows that the identity of connector species in a plant–pollinator network can change drastically across seasons.
Article
Full-text available
Many researchers highlight the role of urban ecology in a rapidly urbanizing world. Despite the ecological and conservation implications relating to carnivores in cities, our general understanding of their potential role in urban food webs lacks synthesis. In this paper, we reviewed the scientific literature on urban carnivores with the aim of identifying major biases in this topic of research. In particular, we explored the number of articles dealing with predation and scavenging, and assessed the geographical distribution, biomes and habitats represented in the scientific literature, together with the richness of species reported and their traits. Our results confirmed that scavenging is largely overlooked compared to predation in urban carnivore research. Moreover, research was biased towards cities located in temperate biomes, while tropical regions were less well-represented, a pattern that was more evident in the case of articles on scavenging. The species reported in both predation and scavenging articles were mainly wild and domestic mammals with high meat-based diets and nocturnal habits, and the majority of the studies were conducted in the interior zone of cities compared to peri-urban areas. Understanding the trophic role of carnivores in urban environments and its ecological consequences will require full recognition of both their predation and scavenging facets, which is especially desirable given the urban sprawl that has been predicted in the coming decades.
Article
Full-text available
Research on human-animal interaction has skyrocketed in the last decade. Rapid urbanization has led scientists to investigate its impact on several species living in the vicinity of humans. Domesticated dogs (Canis lupus familiaris) are one such species that interact with humans and are also called man’s best friend. However, when it comes to the free-ranging population of dogs, interactions become quite complicated. Unfortunately, studies regarding free-ranging dog–human interactions are limited even though the majority of the world’s dog population is free-ranging. In this study, we observed twelve groups of free-ranging dogs in their natural habitat, the streets. We quantified their interactions at the intra (dog–dog) and interspecific (dog–human) levels. The study areas were divided into two zones, namely – intermediate and high flux, based on human activity or movement. Social network analysis revealed higher instances of interspecific than intraspecific interactions, irrespective of the human flux zones. Humans, in significantly higher occasions, initiated both positive and negative behaviors in comparison to dogs. Our findings conclude that humans are a crucial part of the interaction network of Indian free-ranging dogs.
Article
Reproduction provides direct fitness benefits; therefore, it is important to determine why in some societies certain individuals have disproportionate access to reproductive opportunities. The social interactions that underlie reproductive hierarchies can occur in multiple situations, yet they are rarely studied in unison. The reproductive heir in the social wasp Ropalidia marginata is cryptic to human observers until the queen dies or disappears. To determine whether a reproductive heir can be identified through her behaviour, we examined four types of social situations: aggression, spatial overlap, trophallaxis and exchange of solid food. We asked whether accounting for all four social situations in a multilayer network provides more information about the structure of the society than examining each situation on its own or in an aggregate network that does not distinguish between social situations. We found that the reproductive heir had the most social interactions in the multilayer network but not in each of the social situations when considered separately or when all situations were lumped together. Our work demonstrates that multilayer networks can uncover new insights on social organization by explicitly considering the links between multiple situations of social interactions.
Article
Anthropogenic provisioning of food to wildlife is ubiquitous across the globe. It may be intentional such as in the form of bird feeders or offering of food to animals by tourists, or unintentional when animals use anthropogenic food sources like crops, plantations or garbage dumps. Provisioning has profound effects on wildlife ecology and behaviour, but the consequences of these changes to mutualistic and antagonistic interactions have rarely been investigated. Here we demonstrate how provisioning can affect the role of wildlife as seed dispersers. As provisioning directly influences the amount and quality of resources, it can substantially impact fruit removal by wildlife, seed displacement and the probability of germination at the sites of deposition – more often in ways that are detrimental. In most cases, fruit removal decreases as animals choose to feed more on the abundant and higher‐calorie anthropogenic foods. However, for animals whose size and abundance are increased by provisioning, dispersal quantity could be enhanced. Seed dispersal distances are also reduced by provisioning, and seeds are often deposited at sites unsuitable for germination. More studies are required to understand how these broad patterns might vary across different contexts and wildlife species. Provisioning has the potential to greatly modify plant communities through the direct impacts it has on resources and wildlife seed dispersal services.
Chapter
Although the process of converting nutrients sequestered within carcasses has historically been portrayed as occurring at the detrital level, there is a growing consensus that vertebrate scavenging is pervasive among ecosystems across the globe. Throughout this chapter we highlight the central role scavenging plays in ecosystem functions such as nutrient cycling and transport within and among ecosystems, biodiversity maintenance, and disease transmission dynamics. Vertebrate scavenging of carrion also can create more stable food webs by promoting food web complexity, providing routes by which communities may sequester resources at higher trophic levels, and subsidizing populations of vertebrates during periods of food limitation. However, anthropogenic activities that directly alter vertebrate scavenging communities or shift the competitive balance among the various groups of organisms that utilize carrion may have cascading impacts within ecosystems and disrupt ecosystem services provided by carrion consumers.
Article
Human pressure exerts a significant influence on animals and the environment. One of its consequences, plastic pollution is considered one of the major threats to fauna as well as a significant conservation issue. In this research, we examined the global pattern of one example of avian behavior in response to pollution-namely, the incorporation of anthropogenic materials into nests-as well as the existing knowledge on this subject. Based on 25 articles, we studied 51 populations, involving 24 bird species, and checked 10,790 nests; as a result, we found that incorporation of debris is correlated with increasing human influence on the environment, measured as the Human Footprint Index. Moreover, the probability of debris incorporation is higher in terrestrial than in marine species. We also identified knowledge bias in favor of marine as opposed to terrestrial species: namely, marine species attract more scientific attention than terrestrial. Furthermore, research approaches to these two ecosystems differ. Undeniably, the factors which influence debris incorporation by birds, the scale of this behavior, and particular forms of use of debris in bird nests are aspects which require long-term standardized research. This is the first global review paper on debris incorporation by birds to demonstrate a close link to human pressure as a driver.