Transcript Target Groups in Food Webs

- Anusha Uppaluri
Contents
 Problem
 Problems Importance
 Related Work
 Conclusion
 References
 Questions
2
Problem
 Food Webs – Complex networks formed by one species
of animals feeding on the other species
 Target Groups – Species of animals which when extinct
cause the near extinction/ extinction of a large number
of species of animals
3
A Very Simple Food Web
4
Problems Importance
 Food webs make a very complex network and its
understanding would be interesting
 Interdependence of species in food web is important
to maintain balance
 Helps in understanding the do’s and don'ts in trying to
preserve our environment and ecosystem
5
Related Work
•
•
•
•
•
•
•
Dunne et. al.
16 food webs are analyzed
Food webs either have the power law, partial power law,
exponential or uniform degree distribution
Relationship exists between the degree distribution and the
connectance
Connectance – degree to which nodes of a system are
connected to each other
High connectance – uniform degree distribution
Middle connectance – exponential distribution
Low connectance - power law or partial power law
6
 Food webs with low or high connectance aren't usually
found
 Understanding of food web degree distribution and
connectance helps estimating its reaction towards
structural changes
7
Dunne et. al.
 16 food webs were considered
 Primary species loss was simulated
 Robustness of food webs in terms of secondary
extinctions is measured
 Food webs are more robust in case of random species
removal than in case of removal of species with many
links to other species
 Posses thresholds beyond which food webs are quiet
sensitive to highly connected species removal
8
 Food webs with high connectance display high
sensitivity from the beginning
 Food webs displaying power law degree distribution
most vulnerable to attacks
 Over fishing of coastal marine vertebrates have
impacted highly connected species with associated
cascading effects
 Most relevant to the question being focused on
 Talks about how the food web is actually affected by
eliminating certain species
9
Alison J. Gilbert
 Changes in connectance cause changes in robustness?
 If poorly connected species are lost => connectance
increases. Does this => increase in robustness?
 Does converse hold good?
 Robustness – likelihood to lose nodes secondarily
 Fragility - converse of robustness
 Connectance – degree to which nodes of a system are
connected to each other
10
 Loss of poorly connected species – positive change in
connectance
 Loss of highly connected species – negative change in
connectance
 Intercept – number of connections of the species at
which change in connectance becomes positive from
negative
 Intercept is always and integer and loss in species will
lead to change in connectance
11
 Four removal protocols applied to 16 food webs used
by Dunne et. al.
 Four removal protocols – highly connected species
according to connectance, species with least
connectance, Switched for least connected to highly
connected after certain percentage of species have
been lost and species were removed randomly
12
 Steep decline in connectance are more likely to cause
secondary species loss
 Huge negative changes to connectance => large
negative impact on robustness
 Increase in connectance from removal of least
connected nodes => lesser secondary losses but still
reduces robustness
 Loss of any species causes reduction in robustness
irrespective of their connections
 Connectance changes little with random species
removal
13
 Loss of species => change in connectance => loss of
robustness
 Connection between robustness , small negative and
huge positive changes to connectance cannot be
ascertained
 Highly robust food webs => smaller changes to
connectance with species loss
 In food webs with uniform degree distribution, species
loss do not cause large changes to connectance
14
 Robustness is related to degree distribution
 Robustness doesn’t just depend on connectance
 Helps with my research as it answers the question of
how certain species removal might affect
15
Conclusion
•
•
•
•
High connectance – uniform degree distribution
Middle connectance – exponential distribution
Low connectance - power law or partial power law
Food webs are more robust in case of random species
removal than in case of removal of species with many
links to other species
• Food webs with high connectance display high
sensitivity from the beginning
• Loss of any species causes reduction in robustness
irrespective of their connections
16
 Huge negative changes to connectance => large
negative impact on robustness
 Increase in connectance from removal of least
connected nodes => lesser secondary losses but still
reduces robustness
 Robustness doesn’t just depend on connectance
 Robustness is related to degree distribution
17
References
 Dunne J. A., Williams R. J., Martinez N. D., “ Food-web
structure and network theory: The role of connectance and
size”, Proceedings of National Academy of Sciences of
United States of America, Vol. 99 No. 20, pp. 12917-12922,
2002
 Dunne J. A., Williams R. J., Martinez N. D., “ Network
structure and biodiversity loss in food webs: robustness
increases with connectance”, Ecology Letters, Vol. 5, pp.
558-567, 2002
 Gilbert A.J., “ Connectance indicates the robustness to food
webs when subjected to species loss”, Ecological Indicators,
Vol. 9, pp. 72-80, 2009
18
Questions?
19