Geoscience Reference
In-Depth Information
analysis after removing all the empty stomach. In summary, the total length
of individuals capture more variance than the identity of species in the
different temperature and salinity combinations, a result that force us to
identify alternative mechanisms to explain such a large variance in the
number of prey items per individual predator beyond the specific identity
of each species.
D. Expectations from Genetic and Ecological Drift to Study
Diversity in Individual-Based Food Webs
We run Eq.
(26)-(28)
using a sampling size of 10
3
individuals and compare at
steady state the expected with the observed abundance of the 3 and 10 most
abundant prey and predator species, respectively (see
Figure 9
and 'Rarefac-
tion, model analysis and fit to the data'). The model overestimates the less
common fish species in all environmental conditions (
Figure 10
). This is an
expected result because the model did not consider dispersal limitation or
environmental filtering among the sampled locations, a mechanism that limit
the presence of some predator species in some sampled locations. Note,
however, we have analysed just the 10 most common fish species and so
this result suggests that dispersal limitation is strong even for the common
species in the community. Second, the model also consistently underesti-
mates the abundance of fish species with intermediate to high abundance
for the lowest range of salinity at any temperature value and also for the
lowest range of temperature at any salinity value. Predictions of the abun-
dance of fish species in the intermediate range of abundance are better for
larger salinity and temperature values.
Despite likelihood values were consistently similar across all the environ-
mental situations in the fish community (see
Table 1
), the values obtained
from the distributions that best fit the data for the mutation rate and the
minimum genetic similarity value to have fertile offspring fluctuate broadly
across all the environmental gradients sampled. The mutation rate was
correlated with the minimum genetic similarity value to have fertile offspring
(
Figure 11
), but the speciation rate values were independent of the mutation
rate or minimum genetic similarity value.
The observed abundance of the three most common mysid species is within
the confidence interval predicted by the model in most of the environmental
situations, with the exception of low salinity and high temperature with the
second most abundant species outside the confidence interval (
Figure 12
).
Variation among likelihood values was higher than in the fish community.
This is also expected because of the lower number of species in the mysid
community. The mutation rate and the minimum genetic similarity values
