23
8. Pulling It All Together Using Ecosystem Models
A group of scientists led by Dr. Villy Christensen at the
University of British Columbia has taken on the task of
evaluating long-term trends in salmon survival as part
of the Salish Sea Marine Survival Project. The group have
learned a great deal from reconstructing ecosystem history
using complex food web models for about 50 ecosystems
around the world. They concluded that if you wish to
replicate historic trends, you need to understand:
1. the interactions in the food web, including how
predator and prey impacts each other and how their
populations have changed over time;
2. how the environmental productivity, driven by
atmospheric and oceanic conditions, has changed
over time; and
3. how human impacts and impacts on habitat have
changed.
A core aspect of the research is that it requires long-term
data. While the research that is supported by SSMSP
to a large extent is focused on the present, this ecosystem
modelling activity along with the SSMSP data centre
(www.sogdatacentre.ca) gathers past data in an attempt
to understand the reasons for the reduced marine survival
of Coho and Chinook in the Strait of Georgia. But such
historic information is sporadic at best — even with a
40-year time horizon — making it necessary to rely on data
analysis and synthesis to fill in any blanks. In other words:
we need computer models to reconstruct the past, back to
when marine survival of salmon smolts was higher.
Through this initiative, UBC researchers are developing a
coupled hydrographic and biogeochemical model of the
Salish Sea, and linking this to a spatial food web model
in order to evaluate how the combination of changes
in environmental productivity, food web structure and
human impacts (notably through fishing) has changed
in the Salish Sea over recent decades. The overarching
hypothesis is that the environmental productivity of the
Salish Sea is changing over time (e.g., inter-decadal) and
that such changes can be amplified through the food web,
potentially leading to stronger effects on upper-trophic
level species such as Chinook and Coho Salmon.
How do we know if a model is an accurate representation
of the actual events? The best means to test complex
computer models is to actually build similar but
independent models — and that's what we are doing.
Besides the UBC model, the National Oceanographic
and Atmospheric Administration (Seattle, WA) is building
a duplicate model to also explain changes to salmon
productivity in the Salish Sea.
Complementary "end-to-end"
ecosystem models integrate annual
weather and water properties,
food web structure, fish data
and human impacts.
They will be used to cumulatively
assess factors affecting salmon
survival and trends.
Their long-term value may be as
decision support tools for Salish
Sea ecosystem and salmon recovery.
Integrating what we know to help
us manage change over time.
