|
|
Modeling: a logical or mathematical formulation that attempts to simulate some aspect of the real world.
OUTLINE
I. Steps in Cartographic Modeling
II. Advantages and Disadvantages
III. Example
Steps in Cartographic Modeling
1. Statement of Problems or Objectives
2. Statement of Conditions and Assumptions
3. Statement of Methodology
4. Implementation
5. Evaluation
Advantages of Cartographic Modeling
1. Need to define problem clearly
2. Decide on data requirements
3 Flow chart using well defined spatial operations that can be linked
4. Easy to compare a variety of scenarios
Disadvantages
1. Oversimplification of real world
2. Data quality problems
Steps in Modeling (Overton, A Strategy of Model Construction in Ecosystem Modeling: Theory and Practice)
1. Statement of the Objectives and Problems:
State the overall objective of the model. Divide up the problem statement into sub-problems. Unless the objective is specified, there is no direction nor a clear end to the activity. Specification of purpose allows for the transformation from goal statement to objective statement.
Restatement of Problem: Specify the model objectives as a list of model specifications. Is the problem tractable. How can the possible routes to solving the objectives be narrowed.
2. Conditions and Assumptions:
State conditions of problem. This may include the current state, background, or case history of the problem. Assumptions of the model define the limitations of the analysis. Because a model is an approximation of reality, your are always making some set of assumptions when implementing a model. One assumption of most models is that the processes of the past will continue into the future.
Gather information from the scientific literature that is relevant to the objectives or problems the model is attempting to solve. Organize it in manner that is appropriate for addressing the model objectives or problems. If necessary augment this information. Revisit the model specifications in the context of what has been learned. Evaluate the realism of the model.
3. Methodology
Assemblage of the Sub-models into a Model
Sub-problems should be identified and developed as sub-models within the primary model that addresses the larger objectives and problems of the analysis. This will facilitate the development of the model and its evaluation.
a) identification of the sub-problems (or divide and conquer).
b) development of sub-models that addresses the sub-problems
c) develop a strategy for integration of the sub-models
d) develop a flow chart that shows the parts in the context of the whole.
Identify:
a) data sets needed
c) spatial operations
d) non-spatial operations
e) interaction of spatial and non-spatial data
- reclass
4. Implementation
Implement the model using the analytic tools available in the GIS. Are the tools sufficient? Can routes be found to circumvent the limitations of the GIS system?
5. Evaluation
The efficacy of the model should be tested. If the model does not conform to expectations its assumptions and components should be re-examined and adjusted were necessary. The above procedures are performed in an iterative fashion until the objectives are achieved.
Example (hypothetical)
1. Problem: The New York City DEP is interested in measuring the environmental equity (vis a vi economic class) of the siting of waste transfer stations in Brooklyn.
Restatement of Problem: Is one particular income classes bearing the greatest burden of the impact of waste transfer stations
2. Conditions and Assumptions:
a. impact of waste transfer sites on surrounding communities is negligible beyond 1/4
of a mile.
b. within 1/4 of a mile the effect is uniform
c. income classes are distributed evenly throughout a census tract, such that for and
subset of the census tract the tract percentages for each income class will be identical
to the whole.
3. Methodology
a. data sets needed
- income by census
tract
- location of waste
transfer sites
b. spatial operations
- points in polygon
- buffer
- overlay
c. operations on attributes
- select
- reclass
- calculate area
estimates
- generate statistics
4. Implementation
Implement the methodology using your GIS.
Outline for flow of implimentation
- select
out only those waste transfer sites in Brooklyn- generate a 1/4 mile buffer around these sites
- select
income data from the census data and reclass into three income classes: low, medium and high.- relate
to Tiger.pat- add
a dummy area item to save area of each census tract prior to polygon intersection- intersect
Tiger.pat with buffer coverage- recalculate
the income classes based on the percentage of the census tract left in the intersected polygons. Use the dummy item you created that was brought along in the intersection (specified during the intersection).- calculate
the totals for each of three generated income classes for the entire borough of Brooklyn.- generate
pie charts for the number in each income class for: (1) all of Brooklyn and (2) effected areas.- create
map showing output.
5. Evaluation
Develop and discuss the methodology you used for testing your results. Is your model a good one? Does it accurately represent the processes you are modeling? Perform statistical analyses. Include both qualitative and quantitative observations. If not what changes can be made to improve your model. Document those changes and go back up to step 2.
6. Applied to new situation.
- hypothetical transfer waste site.
- income distribution within 1/4 mile of new site.
