Environmental Engineering Reference
In-Depth Information
the forest. This is reflected in great differences in food web structure and the amount
and type of resources produced in both systems. In the Gulf of Nicoya, detritus
matter exported from the mangroves to the estuary feeds an aquatic food web with
shrimps and other aquatic detritivores in the centre of the web, while in the Caet
´
estuary, most energy remains in the benthic domain of the mangrove forest where it
is transferred to an enormous biomass of leaf consuming mangrove crabs, the
principal resource of this system.
5.3.2 Ecosim Approach
The following example is taken from Taylor et al. (2008) who explored the relative
importance of external and internal ecosystem drivers in the Northern Humboldt
Current Ecosystem from 1995 to 2004. External, non-trophically-mediated drivers
considered were changes in phytoplankton biomass, fishing rate (effort and mortal-
ity), and oceanic immigrant biomass (mesopelagic fish).
Figure
5.4
below shows to what degree the time series of biomass of several of
the model groups (open points) were reproduced by the model simulations. Three
Mesozooplankton 200-2000um
Diatoms
*
Dino- and silicoflagellates
*
5
0
Macrobenthos
Sardine - S. sagax
Anchovy - E. ringens
-5
-10
Mesopelagics - Lightfish and
Lanternfish*
Jumbo squid - D. gigas
Horse mackerel - T. murphyi
-15
-20
Mackerel - S. japonicus
Small hake - M. gayi peruanus
Medium hake - M. gayi peruanus
-25
-30
PP
F
I
External Drivers
Large hake - M. gayi peruanus
Seabirds
Pinnipeds
Fig. 5.4
left
: Time-series trends of biomass changes from the data sets (
dots
) and Ecosim
simulations (
lines
). Presented is the best-fit simulation (i.e. lowest SS), using all drivers (phyto-
plankton, PP; fishery, F; and immigrants, I) followed by a “fit-to-time-series” routine. Yearly data
points represent “biological years” (i.e. July-June of the following year).
Asterisks
indicate
artificially-forced functional groups (diatoms, dino- and silicoflagellates, and mesopelagics);
right
: Sum of square (SS) changes under the application of external drivers in all possible
sequences and combinations. All simulations use intermediate, default control settings (i.e. all
predator-prey vulnerabilities equal 2.0). Negative values (i.e. decrease in SS) indicate an improve-
ment in fit (from Taylor et al. 2008)
