Environmental Engineering Reference
In-Depth Information
a
b
c
65
65
65
60
60
60
55
55
55
50
50
50
-10
-5
0
5
10
-10
-5
0
5
10
-10
-5
0
5
10
d
e
f
65
65
65
60
60
60
55
55
55
50
50
50
-10
-5
0
5
10
-10
-5
0
5
10
-10
-5
0
5
10
Fig. 20.2 Particle tracks over the duration of the simulation runs. (a) baseline run; (b-c) runs with
horizontal diffusion with 10 (b) and 100 (c) particles per start location; (d) run with a Gaussian
particle release (spawning) pattern; (e) run with an “ontogenetic” (see text) vertical migration
behaviour pattern; and (f) run with a complex (see text) settlement rule
(Fig.
20.3b
). Mortality slightly smoothed out the juvenile distribution (max. %
within any given grid cell was 7.8 and 3.8 in runs without and with mortality,
respectively), without a dramatic effect in geographical patterns. The largest diffe-
rences were along the areas where juveniles accumulate at the end of the simula-
tions (some of which correspond with actual cod nursery areas), as it could be
expected. The MAE between these 2 runs was 0.0480.
20.3.2 Horizontal Diffusion
Horizontal diffusion increases the horizontal spread of the particle tracks (e.g.
Fig.
20.2b-c
) compared with the purely deterministic tracks of the baseline run
(Fig.
20.2a
). As the number of particles per start location increases, so does the
relative horizontal dispersion of the settled juveniles up to a point when increasing
numbers of particles do not result in significantly different patterns, i.e. the simula-
tion results become independent of stochastic effects. The biggest differences occur
