A look intoDesigning Geodatabases: Case Studies in GIS Data Modelingby David Arctur and Michael Zeiler.

In 2001, ESRI introduced the geodatabase along with ArcGIS 8. The term "geodatabase" is the short form of geographic database and it is fundamentally the data model ArcGIS was developed around. The geodatabase represented the evolution of ESRI's earlier data models, the ArcINFO Coverage and the ArcView Shape file. Unlike its predecessors, the geodatabase offers users the advantage of being able to store data of various types and themes in a single "monolithic" multi-theme (.mdb) file. The geodatabase was offered in two versions: a Personal Geodatabase and the Spatial Data-base Engine (SDE).

The book by ESRI Press titled Designing Databases: Case Studies in GIS Data Modeling, available July 2004, provides some detailed examples of how the geodatabase is used in the real world of everyday GIS operations. It also delves into the theories behind data organization and structure. I would classify it as recommended reading for anyone who is contemplating managing his geospatial data in a GIS format. It is also a useful reference for GIS end users because it provides a detailed overview of the many ways data can be used.

Fig. 1 Survey Analyst Overview.

The Goals of the Publication

The authors state that the three main goals of the book are: 1.) to teach readers practical design concepts; 2.) to present a series of commonly used design patterns for GIS databases; and 3.) to get readers started with geodatabase implementation for projects. They then proceed in their efforts by breaking the process of geodatabase design into 10 steps in three categories.

Conceptual Design

1. Identify the information products that will be produced with a GIS.
2. Identify the key thematic layers based on information requirements.
3. Specify the scale ranges and spatial representations for each thematic layer.
4. Group representations into data sets.

Logical Design

5. Define the tabular database structure and behavior for descriptive attributes.
6. Define the spatial properties of data sets.
7. Propose a geodatabase design.

Physical Design

8. Implement prototype, review and refine a design.
9. Design work flows for building and maintaining each layer.
10. Document a design using appropriate methods.

The authors not only do a good job of illustrating their "10-step method," they avoid the common pitfall of making too many assumptions about their audience. Each detailed illustration includes an information page containing glossary items and a brief yet useful explanation/definition of many of the technical terms used in the book.

The Chapters

Most of the nine chapters are filled with various diagrams and details. I was pleased to see the authors provide quite a bit of detail on three of my favorite issues: topology, building geometric networks, and my absolute favorite, getting accurate with survey data sets. The latter features the Survey Analyst Extension as the preferred method of achieving so lofty a goal.

The chapters Streams and River Networks (Chapter 2), Census Units and Boundaries (Chapter 3), and Addresses and Locations (Chapter 4), are devoted purely to "classic" mapping and GIS activities. But the next two chapters, Chapters 5 and 6, titled Parcels and the Cadastre and Surveying Federal Lands, focus on systems like those many of us work with every day.

Parcels and the Cadastre

The Parcels and the Cadastre chapter of the book is "derived," according to the authors, from the model used by Oakland County, Michigan. Those readers who have visited the Geocom-municator website at www.blm.gov/ nils/general/index.htm, or attended major industry conferences will recognize this schema.

There are excellent framework models for the business process associated with parcel-based cadastres. This chapter enumerates many of the potential uses for parcel-based cadastre and tackles an array of issues regarding the data. But the focus is clearly about the workflow of managing that data.

In addition to the modeling theory and geodatabase structure, there is a section on land and title research. That might seem a bit ambitious or even avant-garde for a GIS tome, but I rather suspect the authors are reaching out to a broader audience.

In the authors' model there is a strong emphasis on using surveys and survey measurements as the core layers for a parcel geodatabase. There is a good overview of the Public Land Survey System and a good bit of detail about how lands are subdivided. And that serves as an excellent segue into the next chapter, Surveying Federal Lands.

Surveying Federal Lands

Again, in Surveying Federal Lands, the word surveying is conspicuous. The chapter opens with a concise but useful history of the Public Lands Survey System and the role of the Bureau of Land Management (BLM).

The authors then go on to describe what they call the "Workflow Process" of creating a database that will produce cadastral plats. Surveyors will be pleased to note there is a good bit of space devoted to detailed explanations of survey source information, measurement and measurement management. The chapter contains numerous examples of table construction (how attribute data actually appears and is accessed in a geodatabase). And flow chart diagrams illustrate design concepts.

The extensive diagrams are taken directly from the NILS (National Integrated Land System) project. The authors explain important nuances like the difference between a corner and a monument, multiple coordinates for a single point, and the importance of the unique identifier for points.

The basis for this robust model was developed from existing BLM sources, and from the Geographic Coordinate Data Base (GCDB) and the Geographic Measurement Management (GMM) system, both of which are familiar to many surveyors.

Additional Resources

The book has a companion website at http://support.esri.com/datamodels. Readers can download several sample data models and white papers from the site.

Other books readers might find helpful to aid in implementing the schema and techniques illustrated in Designing Geodata-bases: Case Studies in GIS Data Modeling are the Dictionary of GIS Terminology, Building a Geodatabase, and Integrating GIS and the Global Positioning System, all published by ESRI Press.

About The Authors (From A to Z)

David Arctur holds a doctorate in urban planning as well as bachelor's and master's degrees in electrical engineering. He is part of the ESRI Geodatabase Development Team, has authored numerous publications and has more than 18 years of programming experience. Michael Zeiler holds an A.B. degree in physics. He has more than 20 years of experience as a software development programmer and has written books on various aspects of geographic information systems for 10 years.

About the Book

Designing Geodatabases: Case Studies in GIS Data Modeling
By David Arctur, PhD, and Michael Zeiler
ESRI Press, Redlands, Calif.
Copyright 2004
ISBN 1-58948-021-X
408 pages