Preliminary scenario results should be available in 6 months.
The web site displays model structure and results and allows people to interact with the broader management/research community.
The model code is publicly available for those wanting to run their own model, however, the model requires very poweful hardware, currently runs only under UNIX, and may be too complex for many people to run.
Plans are being considered that would use the graphic user interface to offer sets of model options (scenarios) that could be controlled and run by an unsophisticated user.
Results could be generated by this spring. In one year from
this summer the model will be about as good as the data allow.
Some questions such as effects of agricultural policies will
be poorly represented by the model at that time because much spatially
explicit agricultural data (e.g. farm practices, profitability)
is kept confidential, preventing estimation of farmer decisions
in response to policy change. Improved data is being developed
and may enhance future estimations of farmer's decisions.
A: Our model differs in two fundamentally different ways:
A: Yes, in the sense that we can set growth levels and
examine changes to ecosystem indicators of health. We can also
assess how policies alter land use distribution on the landscape
which can impact ecosystem health. However, we need to assume
a set of policies in order to test effects and can not work directly
from the goal.
A: The economic model can be implemented for all counties
encompassing the watershed, but would not be valid for other areas.
The ecological model has been applied to other ecosystems, but
the particular findings of the linked ecological economics model
would not necessarily apply to other watersheds. General ecological
or economic modeling techniques which are being developed will
be widely applicable to data in other regions.
A:For one, we could not find an off-the-shelf product that would suit all the needs of our integrated study. Our goal of incorporating land use change within the framework of the model required a fine scale partitioning of the landscape over fairly large areas. At the same time the model had to be simple enough to add other ecological and economic modules without the model becoming unwieldy.
On the other hand, we did use many already existing and
tested hydrologic equations to represent the various processes
in our model (Darcy's law, Manning's equation, etc.). There is
nothing completely new and unproven in the formalism that we apply.
A: Yes it does. Nitrogen is imported into the landscape
model via rain water (wet deposition) and through deposition of
particles (dry deposition). Phosphorous is only imported through
wet deposition. Rates for dry deposition increase when structure
such as buildings or trees are present.
A: Finding the correct parameter values in such a complex
model is a daunting task for a researcher. So, we have developed
a computer-based algorithm which uses a custom
Model Performance Index (MPI).
Basically, the computer does a number of "eyeballing"
runs comparing the available data and boundary or frequency conditions
(e.g. maximum values or seasonal cycles) with the model's output
for different sets of parameters. The parameter values are not
selected randomly (as in a Monte Carlo technique) since the chance
of finding the "good" points in such a big parameter
space would be too small. Rather, the algorithm improves its
search strategy by analyzing the results of each change and running
statistical analyses at every step. This allows us to find acceptable
parameter combinations in a reasonable time, much better than
any human operator or any random algorithm could do. Additionally,
the data gathered to improve the search strategy are of great
value for improving the knowledge of the model's performance,
as well as to quantify the benefit of using additional data which
increases model costs.
A: EPIC is a model specifically designed to characterize
agricultural production. It is more complex and includes many
more parameters related specifically to agricultural land use
than our more generalized unit model could handle given the complexity
of the whole model. Due to EPIC's relatively low error margin
(7-10%) on estimates of crop growth and nutrient uptake rates,
it can be used to calibrate the output of the unit model for agricultural
land use. This enables us to understand the consequences of simplifying
the agricultural model and helps identify ways to reduce error
in the simpler model.
A: A large variety of indices can be calculated directly from model output including ones concerning: hydrologic flow variability, nutrient levels, plant productivity, pattern and proportion of land use, erosion, and general ecosystem health. A linked deer model will be able to estimate indicators related to deer population such as the probability of deer-car encounters.
We are also developing statistical models which link simulation
model output to information about higher trophic levels such as
fish. We hope our continued work with the model will show us
important links in the system which are critical to ecosystem
function and the provision of valuable services.
to submit another question