Environmental Engineering Reference
In-Depth Information
spatial scales, evidence from morphometry, tagging, microchemistry and genetic
studies (Hutchinson et al. 2001; Wright et al. 2006) points at the existence of a cod
meta-population (Hanski and Gilpin 1997), consisting of a number of distinct sub-
populations, relatively isolated, but with some degree of exchange between them.
Therefore, if the spawning biomass of local sub-stocks reaches a low level, where
depensation may threaten recruitment, the extent of immigration from neighbour-
ing sub-stocks may be critical for their recovery. Such exchange may be the result
of the advection of offspring or active migration by juveniles or adults but none of
these processes have been adequately quantified for the species.
Heath et al. (2008) designed an age-structured population dynamics model to
study the consequences of different natal fidelity scenarios on the meta-population
dynamics of North Sea and West of Scotland cod. The model followed discrete
cohorts (fish born on the same date), originating from ten spatially resolved discrete
sub-populations (“natal units”), as their numbers and maturity state (immature or
mature) progressed through time (year-classes). Each sub-population had spatially
defined spawning and nursery areas (identified from field survey data). The propor-
tion of offspring produced in each spawning area which reached a given nursery
area was quantified by Transition Probability Matrices (Paris et al. 2009) calculated
from the output of an offline biophysical model embedded within the population
dynamics scheme. The biophysical model has been described in detail by Gallego
and Heath (2003) and Heath and Gallego (1998, 2000) and it demonstrated that the
offspring transport and survival patterns were not temporally or spatially uniform.
The correct understanding of those patterns could be critical for the preservation of
the cod sub-population structure and potentially for the maintenance of the North
Sea and West of Scotland cod stocks (Heath et al. 2008). To achieve the level of
complexity in the biophysical model appropriate for the goals of this modelling
exercise and the biological patterns that it aimed to represent (Gallego et al. 2007),
I carried out a sensitivity analysis whereby I tested the sensitivity of the model
results (against a baseline model run) to a number of model components of varying
complexity by initially varying one component at time and finally a number of
combinations of those.
20.2 Description of the Analysis
The model made use of daily depth-resolved horizontal velocity fields (resolution
0.25
longitude by 0.125
latitude and 11 fixed-depth vertical layers) generated by a
statistical model of the north Atlantic circulation (SNAC; Logemann et al. 2004)
based on the HAMSOM hydrodynamic model (Backhaus and Hainbucher 1987)
and hourly M
2
tidal velocities. Particles were seeded regularly (at the same hori-
zontal resolution as the HDM) into the model domain within the areas identified as
spawning locations (see above), which were resolved at 1.0
longitude by 0.5
latitude rectangles. For these sensitivity analysis runs, year 2002 flow-fields were
used and the simulations were run from the start of egg production (see below) until
