Geographic dynamics is concerned with the study and application of geographic information from creation through to geocomputation and visualisation. A group from the University Consortium of Geographic Information Science (UCGIS) met with intelligence and non-government participants. That workshop resulted in the book Understanding Dynamics of Geographic Domains, a collection of articles edited by Kathleen Stewart Hornsby and May Yuan. The book includes discussions on the dynamics of change as a means for creating, computing, synthesizing and visualising geographic phenomena.
Understanding Dynamics
of
Geographic Domains
Edited by
Kathleen Stewart Hornsby
and
May Yuan
CRC Press
215 pages; 2008 – ISBN 978-1-4200-6034-8
99.00 USD / 69.00 Euro
Review by Jeff Thurston
Understanding Dynamics of Geographic Domains is a collection of articles arising from a workshop held in 2006 at the Maritime Institute of Technology and Graduate Studies, Linthicum Heights , Maryland. In total 14 geographic information scientists, 25 non-government and 25 U.S. Intelligence Agency representatives with 4 dynamic related education specialists contributed to the workshop. Bringing together the University Consortium for Geographic Information Science (UCGIS) with sponsorship by the Disruptive Technology Office (DTO) of the U.S. Intelligence Service, the book is edited by Kathleen Stewart Hornsby and May Yuan.
The book is divided into 3 parts:
1) Cognitive aspects, representation and data models
2) Analyses, computation and modeling
3) Visualisation and simulation
Contributor Colin Ware asks in the first chapter if we should animate many of the images that cross our paths daily as compared to leaving them as static representations. He reasons that memory processing is impacted when animated content is present and that static images, particularly for new or unknown content which are more easily assimilated when viewed as static images, leading to the conclusion that gathering unfamiliar intelligence is often more effectively processed through static representations.
It is interesting to learn about the processes involved in understanding content and the relationship of memory to represention. Using the example of whale movements, the author describes the use of static images with tagged information as compared to using replay to understand. This simple but effective technique we can often see in split-screen displays, but it is also present in multiple viewpoint displays. I found the text referring to images with color which were not in colour.
Michael F. Goodchild and Alan Glennon point out that the term ‘geographic’ pertains to the surface and near surface of the earth. Furthermore, most geographic dynamics are 3D rather than 2D in nature. By definition the author’s consider geology, criminology, economics and almost all other disciplines to fall within the context of geographic dynamics, not just the study of geography. Accordingly the chapter provides a deeper look into the nature of changing processes as they arise on the earth’s surface.
The limitations of geographic representation author Steven D. Pager indicates, are due to the focus on location and use of Euclidean space. A need exists to advance how we represent reality. This book cites a wealth of past research into the study of representing phenomenon geographically. “The real opportunity for advancement, however, lies with the ability of geographic thinkers to leverage network representations of dynamic geographic phenomena,” Pager says.
In other words, a shift toward the study of geographic processes is suggested, which we are currently seeing in the remote sensing sector and building information modeling sector, both of which integrate operations into workflows and value chains. Further discussion surrounds emerging concepts for studying dynamic networks (complex networks) and their changing interactions. The implications for understanding information networks are wide and varied. The book outlines the applications for studying SARS, airline hubs and highways among others.
In the chapter on exploring the use of gazetteers and their usefulness for analysis and interpretation, author’s Daniel W. Goldberg, John P. Wilson and Craig A. Knoblock disuss the common gazetteers available and identify their relationship to many of the most common web services we now see evolving. In a dynamic and intelligence sense, one might seek to optimise a gazeteer based not only on location, but the prevalence of events or, more specifically, the relatedness of events.
It is important to consider the nature of the data itself and its character as a contributing source with lower and higher levels of uncertainty. Most tools available today, as of this writing, tend to focus on location and coincidence of events. For critical events we still need to develop tools that enable a synthesis of information relative to processes – or those which can create processes, or a hypothesis from data and events. At the present time there is also a trend toward open computing systems with higher levels of interoperability which is enabling higher levels of collaboration, dynamically.
An interesting application involving space-time joining for arsenic levels in a population is provided by author Jaymie R. Meliker. Traditional applications use a ‘what-where’ dyad whereas ‘Time-GIS’ is based upon ‘what-where-when’ triad approaches. Meliker points out that the power of ‘Time-GIS’ applications involves consideration of processes and states, in addition to space-time GIS.
In considering how to explain Time-GIS for readers and its continuous dynamic nature, I would point to the following analogy. Consider projecting one GIS animation on a wall then shining another GIS animation upon it. While we might think of this as a layering exercise (and it is in a sense), the notable difference is that each point can display dynamic information. Now consider shining 15 layers on a wall, each with its own unique information – suddenly each point (pixel) is displaying 17 types of information. Obviously the next question becomes, how much information can one mind and two eyes process, understand and absorb? Yet, we also need to ask ourselves, why represent all information graphically? Why not employ the power of geographic computation, but also utilise embedded intelligence to decipher patterns and so on – automatically? A balance needs to be struck between time, cost, efficiency, capability and result. This applies to both intelligence and consumer applications and remains a growing problem today as the amount of information is growing exponetially.
Mwi-Po Kwan and Fang Ren describe Space-Time behavior and geovisualisation and geocomputation approaches. At this point the book veers decidedly toward 3D environments. The need for increased computing resources is noticeable when speaking about 3D representation since most examples include not only the physical features such as roads and walkways but they also include the movement of objects or people through space, such as the example based on the city of Portland, Oregon. In yet another example, extensibility diagrams are used to display both real and cyberspace representations for indivudals. This brings to mind the application for such work right away. Imagine you want to know where terrorists live and work, but also want to be able to track their online connections and connectivty relative to those changing real locations. This same kind of thinking could be applied to enterprises where one wishes to know about the resources of a manufacturing plant, for example. But to also understand the relationship (and location) of those processes connecting to them – now, in the past or in the future.
The ability to detect change on the landscape using remotely sensed images appears in the mainstream media to be an accomplished fact, but it is not, even today. While it is true that technology has improved, one could quickly argue that much useful intelligence remains undetected in imagery. Furthermore, the connection between detection and visualising large amounts of image related change has not been effectively solved. Nina Lam, Guiyun Zhou and Wenxue Ju discuss these issues in chapter 7 pointing out the need to understand and quantify changes. This involves applying metrics to changes identified. They include a discussion on spatial measures and refer to measures of statistical importance.
The last part of the book is about visualisation and simulation. Narushige Shiode and Li Yin consider spatiotemporal visualisation of built environments. Again, this brings the discussion back around to building information modeling. Three dimensional city modeling is discussed and the author’s consider methodologies for capturing the necessary information for building such models.
In chapter nine, Thomas Butkiewicz, Remco Chang, William Ribarsky and Zachary Wartell discuss visual analysis of urban terrain dynamics. Lidar, SAR and other data gathering techniques are discussed. The concept of merging continuous terrains of multi-resolution is mentioned. However, a deeper discussion of data quality is missing. Having said that, the chapter has direct considerations relating to digital city model development today and describes some of the considerations involved in aggregating data across digital cities where it contains variable resolution.
Some of the issues relating to data storage size and resolution have almost been displaced today by the availability of networked and cloud computing in a utility fashion. Interestingly the ninth chapter includes some truly interesting images which are in colour, unlike most other chapters. In some ways the discussion of digital terrains has been superceded by the fact that we are witnessing the availability of SAR generated high resolution terrain models of homogenous quality across Europe, North America and other parts of the world in 2008. But the inclusion of 3D architecture for Mecklenburg, North Carolina remains as a leading edge example for the integration of buildings with terrain models. A discussion of urban legibility is thought provoking and the images provided, although small provide an understanding of the relationship between urban models and level of detail (LoD).
David A. Bennett and Wenwu Tang write on the topic of mobile, aware intelligent agents in chapter 10. No book on the study of GIScience and dynamics today would be complete without a discussion of agent based modeling and this book is no exception. The interactions between people on the landscape and between the landscape itself in response to changing factors is complex and not fully understood at the present time. There is a difference between ‘contextually aware’ agents and those that are not. The obvious example would include soldiers on a battlefield possessing context awareness of changing situations. Other applications could include vehicles with sensors that are aware of the roads and other vehicles in their path. The nature of such modeling is to increase the ability to make better decisions and to make them quicker.
The final chapter by Narushige Shiode and Paul M. Torrens is on the topic of growth dynamics of real and virtual cities. Metaverses and cyberspaces are discussed and a comparison is made between the virtual city Alphaworld and the real city of Austin, Texas in terms of symmetry and complexity. The value of research within virtual worlds is also mentioned. In considering the comparison, I found myself thinking that it is much easier to navigate a virtual city through Euclidean space than it is to navigate a real city, and that a discussion of growth in urban environments between both must involve consideration of this difference.
In summary I found Understanding Dynamics of Geographic Domains to be an accurate title for this book. It contains a wealth of information about the nature of dynamic geography. The book summarizes much of the previous research leading up to the current state of research and understanding. Although the book would benefit by increasing the size and colouring of many of the graphics, the graphic content supports the text. The value of this book for connecting geography to intelligence gathering operations is fairly straightforward and makes this text useful for that purpose. Beyond that the book identifies many areas of current and future research and will be helpful for people involved in building information modeling, urban planning and the visualisation of environments in 3D space, serving as a good entry to mid-level text on the topic of geographic dynamics.
