Geoscience Reference
In-Depth Information
A
B
3.5
3.0
2.5
2.0
1.5
1.0
100
200
300
400
500
600
100
200
300
400
500
600
C
D
3.5
3.0
2.5
2.0
1.5
1.0
100
200
300
400
500
600
100
200
300
400
500
600
Sample size
Figure 9
Evaluation of the performance of different statistics in the estimation of the
density-mass scaling exponent. Six sample sizes were taken from 100 to 600 indivi-
duals, each resample consisting in 1000 randomly picked individuals from our fish
dataset. Black dots indicate the median values and the dotted line the 95% confidence
interval estimated from resampling. The horizontal dashed line indicates the esti-
mated parameter with all the individuals. (A) Linear and (B) normalized logarithmic
binned histograms; (C) cumulative distribution; and (D) maximum likelihood (which
outperforms all the alternatives).
VI. CONCLUSIONS
In this chapter, we sought to review and advance the theoretical and empiri-
cal determinants of DMR. We consider that a simple ecological pattern, gape
limitation and its effect on prey availability and predation risk, has the
potential to determine the DMR. In the restricted case, where predation
and resources do not change with body size, the energetic equivalence rule
is expected. However, it is predicted that in most cases the balance among the
