Schuurman: Chapter 4

Chapter 4: Bringing it all Together

The real power of the technology emerges from the ability to tell us more about the spatial world than is possible to discern from pieces of data stored in a computer data world.

The power of spatial analysis is exemplified by cadastral systems which are developed and managed by municipalities.
Early GIS was used to call various layers such as property lines, roadways, or common lands up on the screen for visual display.
GIS- different from cartography because of its ability to analyze data.
GIS transforms map data into customized information.
Mathematical theorems helps GIS translate data into relatable information. Example- the Pythagorean theorem is used as a practical way of conveying data for people to see in their everyday lives whether it is at work or at home or on the road.

Users of GIS want to know whether a given spatial entity is contained within an area.
Through the “point in polygon algorithm,” points on a map is tested for inclusion based on the number of vertices between the boundary of the polygon.

Edginess index- shows the shape that has data and how the data is used.
- contributes to spatial understanding
- helps decision making for projects a person may have

Visualization- having an eye for patterns in a set of data. The brain recognizes patterns a lot easier than just regular data without the pattern.

Overlay Analysis, Set Theory, and Map Algebra
Overlay can be used to find the areas of vacant land in a city that are zoned residential and do not cover a shallow aquifer or the number of city parks that lie within a certain residential neighborhood.
The data is examining specific analysis that is the focus of the study, not everything that can be needed.

“GIS enables holistic analysis of geographical problems and is differentiated from other quantitative technologies by the ability to address causal and relational questions at any scale, involving a large number of attributes.” (pg. 93- Dobson)

Buffers are used frequently in GIS to demarcate a zone around a spatial object which should be included in the analysis. Buffering is a way of extending overlay analysis to account for areas that are affected by spatial change as well as to designate protected zones.
Overlays are used for raster data, posing little computational challenge because each layer is already lined-up spatially. In other words, there is no ambiguity within the information spatially on the data set.
Polygon overlay is harder to implement because there is more information that goes into forming the data set such as new geographies.

Set theory: Map algebra is used to calculate new attribute value of population density based on two input layers: area and population.
Categories for this as an instance:
- economical
- affordable
- luxury

Set theory and map algebra are the basis for much modeling and analysis that are developed using GIS. They underlie the structured queries including overlay analysis that planners and modelers use to develop decision-making scenarios and to predict spatial change.

Examples of the environmental model
- Environmental pollution of industrial plants
- Location analysis for creating a landfill for urban waste.
- Explanatory data analysis using visualizations

Spatial Analysis in the Field: Environmental Modeling
A model is important in the field of environmental modeling because they are the best example we have of the interdependencies between the social, cognitive, and technical realms.
Models rely on a morphism or mapping between the entity and the representation.

Each modeling system generates abstract entities and these lines are a form of virtual reality, a rendering. Rendering allows us to find rules for predicting the outcome of experiments in that environment, whether it be the physical area surrounding an industrial park or the rules used by a GIS to predict increases in sea-level.
“All models are wrong, but some are useful.” (pg. 107)

Building Intuitive Models: Multi-Criteria Evaluation (MCE)
MCE- A raster based modeling tool that allows users to combine several attributes in order to derive a suitability index for location on a spatial entity.
Steps in MCE
- Define the problem and relevant criteria.
- Each criteria is scored depending on its relevance to the spatial solution.
MCE is based off factors and constraints- both are relevant criteria.
Constraint- a Boolean criteria that limits analysis to particular geographical areas.
Factors- criteria that define the degree to which a region is suitable.
Factors are conditions that influence the suitability of a given piece of land while a constraint is a condition that limits the suitability of alternatives.
Both deal with multiple perspectives and the results of analysis vary depending on how factors and constraints are ranked in relation to each other.
- Do they satisfy each other?
- Do they satisfy the conditions that will eventually make the environment of the data?
- Do they make the known data set stable?

Brian Harley- “the map is not the territory.”
Models are not that either. They are a way of simplifying representation so that we can better interpret the viability of environments for specific applications. The problem with models is that they are confused with reality when, in fact, they are complex systems subject to change with modification of every factor and constraint.

The Power of the Eye: Visualization and the New Cartography
Visuality and “intuition” differentiate GIS from the approach to geography that dominated the quantitative revolution. Intuition- used as a means of making sense or interpreting visual displays of geographical data.

Exploratory Data Analysis (EDA) and Knowledge Discovery (KDD)- relies upon assumptions of direct correlation between visualization and communication of information. People use visualization as a means to interpret visual imagery, algorithms for data manipulation and patterns of human-computer interaction.

Scientific visualization- a combination of simulation, data analysis, and visualization of complex scientific relationships such as chromosomal structures and ecological dependencies.

Examples:
- genetics
- ecology
- atmospheric science

EDA process and stages:
1. exploratory visualization using a GIS
2. statistical and numerical analysis of spatial patterns and trends (factor and cluster analysis, in this case)
3. visualization and communication of results using a GIS
4. repetition of stage 2 if needed with different statistical and or numerical methods.
5. final results as tables and graphs, maps of diseased clusters or scenarios of disease propagation.

From Data to Analysis: A Case Study of Population Health
By illustrating the population health, it shows the data compromised by assumptions of homogeneity associated with vector polygons, a common theme for vector problems through GIS.

Raster data is better because you can integrate multiple datasets (sets of information) and better facilitate the “data speaking for themselves.”
Deals with the modifiable area unit problem (MAUP); a common problem in GIScience.
Problem occurs when spatial units such as postal codes or enumeration areas are aggregated into larger units, or when a map is redrawn at the same scale using a different spatial division.

Another problem: ecological fallacy
In which aggregation or scaling produces a bias when attributing the characteristics of populations or groups to individuals.
Dasymetric mapping uses local knowledge to assist in areal interpolation. Familarity with the region combined with the use of high-resolution air-borne imagery, in this case, was used to identify areas with known high or low population densities.

MCE and Analysis
Jarman 8- Scale to measure the underprivileged associated with a particular spatial area. Used for research allocation and health care planning. Broken into 8 classes that affect their workload in the UK.

From a GIS perspective, one way of allowing the data to speak is to use raster queries to isolate clusters of like variables that are associated with particular health outcomes. To that end, a motivating questions must be translated into factors and constraints for use with MCE.
You could standardize the data as well to get a better look at the range of the data.

Calculation and the Rationalities of GIS
Michael Curry- GIS clearly supports a trend towards a more surveillant society…Adding a spatial component to databases allows users to locate and describe individuals rather than demographic clusters.
Jeremy Crampton- practices associated with a particular time and place are reflections of larger ways of thinking- rationalities.
GIS’ calculating methodologies are servants of social goals.

2 Comments

Comments RSS TrackBack Identifier URI