model should demonstrate how the system works, with particular emphasis on anticipated system responses to stressor input. The model also should
indicate the pathways by which the system accommodates natural disturbances and how the system may acquire resilience to disturbance. These
processes could be portrayed by illustrating the acceptable bounds of variation of system components and the normal patterns of variation in input
and output among the model elements (Noon et al. 1999). Typically, that
framework is grounded in existing data from either direct observation,
extrapolation from similar studies, or ecologically sound assumptions. These
data may include biological data about the organism and/or remotely
sensed data about the habitat as well as other biological and abiotic data
specific to the question. The conceptual model must address the question
while recognizing data needs and availability to construct, verify/calibrate,
test, and run the model. Once the conceptual model has been formulated,
data needs can be identified and prioritized, and the process of data collection and model construction can begin.
10.1.3 Model Formulation and Construction
The conceptual model is then formalized into a set of mathematical expressions or fuzzy logic and algorithmic steps. The data required for this process
typically pertain to parameter values for the mathematical expressions and
related geographic information system (GIS) data.
The growing use of spatially explicit models, usually based on GIS,
has led to a profusion of models in which observations of existing populations and communities are used to infer relationships among various geographic data sets and habitat requirements [e.g., Akcakaya and Atwood
(1997); Gerrard et al. (2001); Knick and Dyer (1997)]. Other spatially
explicit models, [e.g., Mann et al. (2000)] make predictions based on parameters that are determined from the biology of the organisms or communities. Parameterization of these models is typically based on data from
literature surveys or assumptions from ecological theory. Construction
and use of GIS-based models, of course, requires geographic data sets.
While the use of spatially explicit models may complicate model building,
it can lead to breakthroughs in large-scale understanding or the incorporation of social and economic factors into previously limited analyses
(D’Erchia 1997).
10.1.4 Model Verification and Calibration
Verification is the process of confirming that the model performs as
expected. The data required to verify the model must include a reasonable
sample of input data and an associated set of experimental results or observations to compare with the model output. Calibration can generally be
182
David Hohler et al.
indicate the pathways by which the system accommodates natural disturbances and how the system may acquire resilience to disturbance. These
processes could be portrayed by illustrating the acceptable bounds of variation of system components and the normal patterns of variation in input
and output among the model elements (Noon et al. 1999). Typically, that
framework is grounded in existing data from either direct observation,
extrapolation from similar studies, or ecologically sound assumptions. These
data may include biological data about the organism and/or remotely
sensed data about the habitat as well as other biological and abiotic data
specific to the question. The conceptual model must address the question
while recognizing data needs and availability to construct, verify/calibrate,
test, and run the model. Once the conceptual model has been formulated,
data needs can be identified and prioritized, and the process of data collection and model construction can begin.
10.1.3 Model Formulation and Construction
The conceptual model is then formalized into a set of mathematical expressions or fuzzy logic and algorithmic steps. The data required for this process
typically pertain to parameter values for the mathematical expressions and
related geographic information system (GIS) data.
The growing use of spatially explicit models, usually based on GIS,
has led to a profusion of models in which observations of existing populations and communities are used to infer relationships among various geographic data sets and habitat requirements [e.g., Akcakaya and Atwood
(1997); Gerrard et al. (2001); Knick and Dyer (1997)]. Other spatially
explicit models, [e.g., Mann et al. (2000)] make predictions based on parameters that are determined from the biology of the organisms or communities. Parameterization of these models is typically based on data from
literature surveys or assumptions from ecological theory. Construction
and use of GIS-based models, of course, requires geographic data sets.
While the use of spatially explicit models may complicate model building,
it can lead to breakthroughs in large-scale understanding or the incorporation of social and economic factors into previously limited analyses
(D’Erchia 1997).
10.1.4 Model Verification and Calibration
Verification is the process of confirming that the model performs as
expected. The data required to verify the model must include a reasonable
sample of input data and an associated set of experimental results or observations to compare with the model output. Calibration can generally be
182
David Hohler et al.