• Interpretation and refinement of the information
• Development of recommendations for watershed protection for decision
makers
The process whereby questions are asked, data are collected and analyzed, and decisions made can go askew at many junctions. At the USEPA,
we have experienced several common pitfalls and see several common
areas to be addressed:
• Misguided conceptualization can result in imperfect or wrong questions,
and therefore appropriate models are not developed.
• Existing data may be used in place of appropriate data.
• Data are misused.
• Scale and statistical validity are ignored.
• Inadequate communication is practiced.
9.3 Model Conceptualization
The question and model-conceptualization phase becomes imperfect for
numerous reasons. The basic definition of a model is a simple representation of the relationship of items in the real world (see Chapter 1, this
volume). Models are used to simplify the real world and make it easier to
understand. The endpoint of many models is not the support of decision
making but clarifying complex relationships. Models often neglect major
parts of systems (basically linearizing them in the time–space domain of the
model) to achieve this simplification.
Solutions to environmental problems depend on the questions being
asked, available data for developing the model, and conceptual framework
built into the model. Conversely, existing data often frame the model and
the questions that can be asked (Figure 9.2). Often, a mismatch occurs
between the questions that need to be answered and the data that exist to
answer the question. Often the data that are available to the modeling
process do not directly fit the model. It may be temporally irrelevant, spatially inappropriate, or acquired for a different purpose. The existing data
may be able to answer other questions than the desired ones. The solutions
are to (1) reframe the questions to fit existing data, (2) acquire more data
that are relevant to the questions, (3) reframe the questions and acquire
more data, and (4) find other tools or models that can integrate the needed
questions and existing data. It is often necessary to iterate through this
process to develop optimum solutions to the questions being asked. In a
recent volume edited by Peine (1999) several authors discuss how existing
data were used to conduct an environmental assessment, including limitations dealt with, models used, and potential applications of the information
that was developed. In times of receding budgets for data gathering, the use
of existing data may become even more important.
170
John R. Richardson and Cory W. Berish
• Development of recommendations for watershed protection for decision
makers
The process whereby questions are asked, data are collected and analyzed, and decisions made can go askew at many junctions. At the USEPA,
we have experienced several common pitfalls and see several common
areas to be addressed:
• Misguided conceptualization can result in imperfect or wrong questions,
and therefore appropriate models are not developed.
• Existing data may be used in place of appropriate data.
• Data are misused.
• Scale and statistical validity are ignored.
• Inadequate communication is practiced.
9.3 Model Conceptualization
The question and model-conceptualization phase becomes imperfect for
numerous reasons. The basic definition of a model is a simple representation of the relationship of items in the real world (see Chapter 1, this
volume). Models are used to simplify the real world and make it easier to
understand. The endpoint of many models is not the support of decision
making but clarifying complex relationships. Models often neglect major
parts of systems (basically linearizing them in the time–space domain of the
model) to achieve this simplification.
Solutions to environmental problems depend on the questions being
asked, available data for developing the model, and conceptual framework
built into the model. Conversely, existing data often frame the model and
the questions that can be asked (Figure 9.2). Often, a mismatch occurs
between the questions that need to be answered and the data that exist to
answer the question. Often the data that are available to the modeling
process do not directly fit the model. It may be temporally irrelevant, spatially inappropriate, or acquired for a different purpose. The existing data
may be able to answer other questions than the desired ones. The solutions
are to (1) reframe the questions to fit existing data, (2) acquire more data
that are relevant to the questions, (3) reframe the questions and acquire
more data, and (4) find other tools or models that can integrate the needed
questions and existing data. It is often necessary to iterate through this
process to develop optimum solutions to the questions being asked. In a
recent volume edited by Peine (1999) several authors discuss how existing
data were used to conduct an environmental assessment, including limitations dealt with, models used, and potential applications of the information
that was developed. In times of receding budgets for data gathering, the use
of existing data may become even more important.
170
John R. Richardson and Cory W. Berish