Good Scholarship with GIS Models

The diagram below is an example of what Patrick Geddes called a Thinking Machine. The modeling terms given in the diagram have relationships with eachother in several dimensions, which will become intuitively unforgettable once we discuss a few example models using these terms. This frameworkprovides the critical phases of representing places and spatial processes with data, maps and models, and for understanding how a map or model can be successful, or how to describe effectively, how it should be improved.

In lecture, the relationships among these concepts may be demonstrated with one of the following sample datasets:

Purposeful Use of Information

We seek to answer useful questions by combining information with ideas. At the same time we are aware that some answers may be useful while other answers may be rubbish. We have to learn to judge one from the other. Once we have informed ourselves, we must communicate with others. Understanding the details of this is what makes a productive, resourceful scholar in the digital age.

Questions and Purposes

A clear statement of Purpose is key to understanding a map or a model. Our purpose for modeling may be to explore a question about the way a place preforms, or we may want to generate some new information that will provide needed background for making a decision. In an academic setting our purpose may be to learn something about the suitibility of specifric datasets for specific applications and to learn something about the ways that models work (or fail). Defining the purpose of an investigation is an art that can make your task reasonable and interesting or foolish and impossible waste of time.

Purposeful Abstractions

It is good to be able to organize ideas and facts. These are the raw material of understanding. An example of a fact may be that there are wetlands located within our area of interest. An example of an idea may be that wetlands may be affected by land that is nearby. This logical chain of facts and ideas is a Conceptual Model. The conceptual model can be decomposed into individual concepts, which may be facts or concepts of association. In our simple model, one of our concepts is a fact: Wetlands; and another is an association: Land Near Wetlands. The factual components of a conceptual model may be represented with Data and the associative ideas may be represented with software Procedures including graphic portrayal or other more complex procedures. Data Models are the product of using data and software to instantiate a conceptual model. A model may be a simple portrayal of a single dataset as a map, or it may be more complicated. By adjusting the data and associations in a data model we may investigate experiment with new ideas that may serve to improve our understanding of how places work under different conditions.

Critical Evaluation

As we strive to develop our understanding, we should be mindful of the fact that the world is a very complex and dynamic place. Any facts that we try to represent with data are likely to be incomplete, imprecise, and to a degree erroneous. The software procedures that we use to simulate associations among facts are very rarely perfect simulations of relationships in the real world. So we can comfortably say that all data and all models are inaccurate. The more important question is whether our data and our procedures are accurate enough and this question can only be answered with reference to a specific purpose. Our stated purpose provides the context for evaluating whether our data and procedures reflect the concepts in our model well enough, and the nature the errors we should expect in the key components and how these would propogate as we transform and derive new information. Ultimately we will judge the model or map according to its utility for our developing a Useful Understanding, which might be qualified with aome level of confidence.

Keep in mind that a each dataset that we employ constitutes a model in itself -- with its own independent purposes and methodologies. Each of these needs to be understood through its Metadata (the data about the data) to understand the purposes and the methodologies behind its origin -- in our critique.

Hueristic Refinement of Questions and Models

The process of developing models and questions should be thought of as more cyclical than linear. It is usually the case that the questions and conceptual models that we first concieve are easily shown to be too simple or too complicated once we begin the mental exercise of trying to think of data and procedural proxies for the concepts in our model. We also can do a substantial amount of critical evaluation of these components before we even get started in the act of developing a data model. Often the construction of a model that do not create any useful understanding, is a sign that the process of reworking the question and conceptual model in light of available data and practical procedures has not been given enough time an attention. For example, after working through our model of the interaction of wetlands and the areas near-by, we are likely to find that there are many different conceptions of Near that should be evaluated, and this will require us to refine our conceptual model to add a concept describing the actual assocotiation of interest, such as runoff from pavement and fertilized areas.

Mapping and Experimental Models

So far, all that we have discussed applies to conceptual models that may be addressed with a aspect of a dataset, portrayed and discussed in its topographic context. Many of the associative aspects of conceptual models can be represented with simple graphic visualization. Other types of relationships may involve procedures that transform and create new data -- not just pictures. In these cases, it may be possible to perform experiments by running the model different ways: We may perform Calibration Runs by running the model with varied procedures. We may investigate Alternate Scenarios by making new versions of critical datasets to represent different real world conditions. Or we may do Sensitivity Analysis by making model runs model with alternate data sources representing the same concept in our model. These are very useful to do if you would like understand the impact of using a lower quality (but cheaper and more widely available) dataset. These experimental models can be a lot of fun to play with -- they can be of great practical use in planning and decision support, and they also give us a whole new dimension to work with in our understanding of the modeling activity itself. When you go beyond maps and one-off models, you should challenge yourself by playing investigating how your models behave with different settings or in investigation of alternative futures or sensitivity analyses.

Understanding

At the highest level, our purpose is to increase in understanding. Once we have arrived at an interesting conceptual model, represented it in a data model and critiqued the parts and the whole, then we should at least be able to speak of our improved understanding of Data Needs If we could make recommendations on how better data could be found or made, this would be a good contribution for those that succeed you. You should also now be in a position to your discussion of Simulation Challenges. In a sense, this will return the question of how well your conceptual model covers the critical relationships and process that act in the real world, and also how well the information systems that you have are able to represent simulate these. What are your recommendations should this study be repreated by someone who may heva access to better resources (or more time) than you have? Ultimately your map or model be useful, either for understanding some aspect of the world or of conceptual models and the neaed for information and simulation, or it is simply not worth discussing.

Communication and Credibility

If we have learned something useful by exploring data and models, then our next step is to communicate our understanding to others. Communication is a powerful ability to create understanding in the minds of other people. There are three potential outcomes of an act of communication: We hope that the beholder of our presentation believes that our understanding is credible and our mission is accomplished. Alternatively, because of lack of clarity, our audience may lose interest or become distracted from the points we are trying to make. Worst of all, we may very effectively (but unintentionally) communicate a message that we have misunderstood our subject and are in fact, deluded. All three of these outcomes represent stages along a spectrum of credibility. These are developed by employing effective techniques of portrayal and documentation of information, and also by selective description of the most important elements of your critical evaluation of your model's components.

Replicability and Stewardship of Information

In many approaches to scientific and scholarly work, the notion of replicability is an important concern. Even in everyday wotk, the resources and the procedures that we assembled to support some sort of argument should be set up and documented in such a way that is is repeatable. This is a good thing for us, since models and maps are seldom right the first time, and the ability to easily go back and tweak them later without starting over from scratch. Organizing work with replicability in mind allows us to participate in a much higher level of scholarship, where our work can be built upon by others. When more people understand this, we will see the shaping of an open source laboratory for experimenting with the world! Replicability requires that we think about the organization of the data and the procedures and the documentation that is necessary to understand it. IN some sense, this is a matter of being neat about filing things. In another sense, this requires some formal understanding about how the compoents of our information stick together or not.


	Geographic Information Systems (+/-) Geographic Information Systems in the Design School Getting Started With ArcGIS at the GSD GIS Software Manuals GIS Courses at the GSD Links for GSD GIS Fans Data Resources (+/-) GIS Data Resources GSD Data Collection Downloading US Census Data: 2000 - 1970 US Business Directory Digital Elevation Models Sources of Geographic Images Data Handling (+/-) Understanding GIS Data, Metadata and Schema ArcMap 101: Exploring Peoria Data Beginning a GIS Database Concepts of Geographical Referencing Systems Why Don't My Layers Line Up? AKA: ArcMap Projections Tutorial Georeferencing Scanned Maps and CAD Data Reclassifying Data with Lookup Tables Converting GIS Layers to DXF Geocoding Clipping an Ocean Polygon Effective Cartography (+/-) Elements of Cartographic Style Basic Cartography in ArcMap Mapping with Quantitative Data About Mapping Census Data Mapping Census Data Tutorial Analytic Techniques (+/-) Modeling What's Important A Development Buildout Model Real Estate Tutorial Guidelines for Documenting Models Mastering Relational Procedures in ArcMap Vector GIS Procedures in ArcMap Vector GIS Procedures in ArcMap (08 Version) Raster GIS Procedures in ArcMap Classifying Satellite Images Designing for Visibility Zakem Bridge View Corridors Evaluating Development Scenarios Topographic Modeling in 3D (+/-) Experimenting with Zoning Dimensional Regulations in 3d A rough Understanding of 3D Site Context with ArcGlobe Graphics and Animation in ArcGlobe Basic Terrain Modeling (ArcGIS) Experimenting with Massing Scenarios in SketchUp Beginning a 3d Site Model in SketchUp Good Site Model Structure in Sketchup Interchanging 3D Data between ArcGIS and 3dStudio Topographic Modeling in FormZ (+/-) FormZ 101: A Field of Dreams Modeling Terrain with Contours and Surfaces (formZ) Modeling Surfaces with Mapped Images (formZ) Analytical Graphics from 3-D Site Models (formZ) Metropolitan Scale 3d Models (+/-) The Boston Virtual City Collective Retrieving Scenarios from the Metropolitan Model Repository Creatning Historic and Future Scenarios Studying Urban View Corridors Integrating Administrative CAD Data, GIS and Google Earth Export NYC Data to Sketchup
				Good Scholarship with GIS Models The diagram below is an example of what Patrick Geddes called a Thinking Machine. The modeling terms given in the diagram have relationships with eachother in several dimensions, which will become intuitively unforgettable once we discuss a few example models using these terms. This frameworkprovides the critical phases of representing places and spatial processes with data, maps and models, and for understanding how a map or model can be successful, or how to describe effectively, how it should be improved. In lecture, the relationships among these concepts may be demonstrated with one of the following sample datasets: A couple of Models for determining areas of concern related to stormwater runoff. A model for analysis of Urban Open Space Supply and Demand. A model for experimenting with the impact of alternate zoning regulations Purposeful Use of Information We seek to answer useful questions by combining information with ideas. At the same time we are aware that some answers may be useful while other answers may be rubbish. We have to learn to judge one from the other. Once we have informed ourselves, we must communicate with others. Understanding the details of this is what makes a productive, resourceful scholar in the digital age. Questions and Purposes A clear statement of Purpose is key to understanding a map or a model. Our purpose for modeling may be to explore a question about the way a place preforms, or we may want to generate some new information that will provide needed background for making a decision. In an academic setting our purpose may be to learn something about the suitibility of specifric datasets for specific applications and to learn something about the ways that models work (or fail). Defining the purpose of an investigation is an art that can make your task reasonable and interesting or foolish and impossible waste of time. Purposeful Abstractions It is good to be able to organize ideas and facts. These are the raw material of understanding. An example of a fact may be that there are wetlands located within our area of interest. An example of an idea may be that wetlands may be affected by land that is nearby. This logical chain of facts and ideas is a Conceptual Model. The conceptual model can be decomposed into individual concepts, which may be facts or concepts of association. In our simple model, one of our concepts is a fact: Wetlands; and another is an association: Land Near Wetlands. The factual components of a conceptual model may be represented with Data and the associative ideas may be represented with software Procedures including graphic portrayal or other more complex procedures. Data Models are the product of using data and software to instantiate a conceptual model. A model may be a simple portrayal of a single dataset as a map, or it may be more complicated. By adjusting the data and associations in a data model we may investigate experiment with new ideas that may serve to improve our understanding of how places work under different conditions. Critical Evaluation As we strive to develop our understanding, we should be mindful of the fact that the world is a very complex and dynamic place. Any facts that we try to represent with data are likely to be incomplete, imprecise, and to a degree erroneous. The software procedures that we use to simulate associations among facts are very rarely perfect simulations of relationships in the real world. So we can comfortably say that all data and all models are inaccurate. The more important question is whether our data and our procedures are accurate enough and this question can only be answered with reference to a specific purpose. Our stated purpose provides the context for evaluating whether our data and procedures reflect the concepts in our model well enough, and the nature the errors we should expect in the key components and how these would propogate as we transform and derive new information. Ultimately we will judge the model or map according to its utility for our developing a Useful Understanding, which might be qualified with aome level of confidence. Keep in mind that a each dataset that we employ constitutes a model in itself -- with its own independent purposes and methodologies. Each of these needs to be understood through its Metadata (the data about the data) to understand the purposes and the methodologies behind its origin -- in our critique. Hueristic Refinement of Questions and Models The process of developing models and questions should be thought of as more cyclical than linear. It is usually the case that the questions and conceptual models that we first concieve are easily shown to be too simple or too complicated once we begin the mental exercise of trying to think of data and procedural proxies for the concepts in our model. We also can do a substantial amount of critical evaluation of these components before we even get started in the act of developing a data model. Often the construction of a model that do not create any useful understanding, is a sign that the process of reworking the question and conceptual model in light of available data and practical procedures has not been given enough time an attention. For example, after working through our model of the interaction of wetlands and the areas near-by, we are likely to find that there are many different conceptions of Near that should be evaluated, and this will require us to refine our conceptual model to add a concept describing the actual assocotiation of interest, such as runoff from pavement and fertilized areas. Mapping and Experimental Models So far, all that we have discussed applies to conceptual models that may be addressed with a aspect of a dataset, portrayed and discussed in its topographic context. Many of the associative aspects of conceptual models can be represented with simple graphic visualization. Other types of relationships may involve procedures that transform and create new data -- not just pictures. In these cases, it may be possible to perform experiments by running the model different ways: We may perform Calibration Runs by running the model with varied procedures. We may investigate Alternate Scenarios by making new versions of critical datasets to represent different real world conditions. Or we may do Sensitivity Analysis by making model runs model with alternate data sources representing the same concept in our model. These are very useful to do if you would like understand the impact of using a lower quality (but cheaper and more widely available) dataset. These experimental models can be a lot of fun to play with -- they can be of great practical use in planning and decision support, and they also give us a whole new dimension to work with in our understanding of the modeling activity itself. When you go beyond maps and one-off models, you should challenge yourself by playing investigating how your models behave with different settings or in investigation of alternative futures or sensitivity analyses. Understanding At the highest level, our purpose is to increase in understanding. Once we have arrived at an interesting conceptual model, represented it in a data model and critiqued the parts and the whole, then we should at least be able to speak of our improved understanding of Data Needs If we could make recommendations on how better data could be found or made, this would be a good contribution for those that succeed you. You should also now be in a position to your discussion of Simulation Challenges. In a sense, this will return the question of how well your conceptual model covers the critical relationships and process that act in the real world, and also how well the information systems that you have are able to represent simulate these. What are your recommendations should this study be repreated by someone who may heva access to better resources (or more time) than you have? Ultimately your map or model be useful, either for understanding some aspect of the world or of conceptual models and the neaed for information and simulation, or it is simply not worth discussing. Communication and Credibility If we have learned something useful by exploring data and models, then our next step is to communicate our understanding to others. Communication is a powerful ability to create understanding in the minds of other people. There are three potential outcomes of an act of communication: We hope that the beholder of our presentation believes that our understanding is credible and our mission is accomplished. Alternatively, because of lack of clarity, our audience may lose interest or become distracted from the points we are trying to make. Worst of all, we may very effectively (but unintentionally) communicate a message that we have misunderstood our subject and are in fact, deluded. All three of these outcomes represent stages along a spectrum of credibility. These are developed by employing effective techniques of portrayal and documentation of information, and also by selective description of the most important elements of your critical evaluation of your model's components. Replicability and Stewardship of Information In many approaches to scientific and scholarly work, the notion of replicability is an important concern. Even in everyday wotk, the resources and the procedures that we assembled to support some sort of argument should be set up and documented in such a way that is is repeatable. This is a good thing for us, since models and maps are seldom right the first time, and the ability to easily go back and tweak them later without starting over from scratch. Organizing work with replicability in mind allows us to participate in a much higher level of scholarship, where our work can be built upon by others. When more people understand this, we will see the shaping of an open source laboratory for experimenting with the world! Replicability requires that we think about the organization of the data and the procedures and the documentation that is necessary to understand it. IN some sense, this is a matter of being neat about filing things. In another sense, this requires some formal understanding about how the compoents of our information stick together or not.