Geoscience Reference
In-Depth Information
In this study, we explore the sensitivity of the species-level connectance to
variation in the diets of individuals by studying the effect of individual
sampling effort at the intraspecific level. The fraction of possible links in
the species-based food web is defined as the connectance, C
L/S 2 , where L is
the number of trophic links and S is the number of species in the food web.
The species-level matrix has size, A 13,10 (i.e. 10 most abundant fish prey, 3
mysid prey and 10 most abundant fish predators).
In the species-level food web, a ij ¼
¼
1ifpreyspeciesi and predator species j
interact. We now consider how to aggregate individual-level observations to
determine whether prey species i and predator species j interact. We first study
connectance given by the 'minimal sampling threshold': a ij ¼
1 if at least 1, 2,
3,
, N individual predators of species j have at least 1 individual prey of species
i. We then consider how a tiny increment in the sampled prey individuals within
the stomach contents of each predator alters connectance. Now a ij ¼
...
1ifatleast
1, 2, 3,
, N individual predators of species j sampledhave at least 2 (insteadof 1
as in the first case) individual prey of species i. We compared the neutral case for
which per-individual predation rates are equivalent with the empirical values.
...
III. RESULTS
Figures 5-13 summarize the model's predictions for the number of individual
preys per predator ( Figures 5 and 6 ), the number of prey items as a
function of predator length (Figures 7 and 8) diversity of fish and mysids
( Figures 9-12 ) and the effect of intraspecific variability and sampling effort
on species-level food web connectance ( Figures 13 and 14 ). In summary, we
show that the extremely high variability in individual's number of prey at
intraspecific and interspecific level has dramatic consequences for the
connectance of the food web at the species level. This variability significantly
departs from the expectations of the eco-evolutionary model in all the
environmental conditions. Despite total length of individuals captures
more variance in the number of observed prey per individual predator than
the identity of species, this variance represents only a minor fraction and it
remains consistently the same across the different temperature and salinity
combinations. Finally, for the parameter combination explored, the eco-
evolutionary dynamics model predicts broad fluctuations in the speciation
rate that best predict the data and lower than expected fish diversity in parts
of the salinity and temperature gradients.
A. Sampling Effects in Individual-Based Food Webs
Twenty years ago, Gary Polis emphasized several limitations in the food
web data ( Polis, 1991 ). Among others, 'inadequate dietary information'
was one of his main concerns. Because the number of prey items recorded
Search WWH ::




Custom Search