Cartographic Abstraction

Purpose of lecture:

More points about Geographic Data

Muehrcke's Chapter 2 is full of random snippets (as is chapter 3, as well). A few of these didn't fit the theme last time, but bear some recognition.

Sources of Data (Acquisition Methods):

• Ground Survey (we will deal with this under distance and direction, lecture 11)
  • This section actually spends much more time on reference systems (lecture 10)
• Census
  • (deals with shape, size and orientation of census units)
  • It is really an issue of assigning attributes to objects.
• Remote Sensing (see lecture 7)
• Compilation (use of secondary sources (reuse of primary materials); backbone of cartography)

"Inventory Scheme":

This is a rather odd title, it really belongs with the acquisition methods, somehow)

• Population counts (exhaustive sources)
• Spatial samples
  • This fits into my discussion of surfaces. You really cannot get all the values of a continuous surfaces (there are an infinity (or actually a two-dimensional infinity) of them... Sampling is usually not so studiously designed as this section pretends. Our precipitation stations are at airports were someone can maintain them.

"Spatial Prediction":

This is another odd title, I would call this transformations

• point to point conversion (introduces interpolation and extrapolation)
• cartographers do have to make up the data...
• area to point (geometric simplification); area to area (overlay, see lecture 17, 18)

Derived values

This section introduces a lot of issues that we will cover later...

Cartographic Abstraction

Muehrcke lists five processes:

Selection

This is a grab-bag of items, some of which are vaguely related to abstraction, and some of which are more tangential (or just not a good fit).

Region, time frame and "variables to be mapped" are the components of time, space and attribute. What doesn't fit as well is the "selection", these three are not really treated equivalently...

Scale is mentioned (see lecture 10), then it spends six pages on "perspective". Lecture 12 and the lab will deal with projections, the rest about oblique perspective might be used for ski areas, but very little else.

Classification

Here, the cartographic abstraction begins in earnest.

As described in the last lecture, categories are the basis of objects, and thus a lot of mapping.

The main question should deal with

• how objects are grouped together
• surfaces are "sliced" into contours

BUT, Muehrcke deals primarily with a thematic problem of

Selecting Class Intervals

This is the situation of having a number (interval/ratio) for a discrete set of objects (eg. counties)

First choice: how many classes

Second choices: where to set the class breaks

The risks:

The normal approaches

This topic is treated in more depth in Geog 360, as a part of making maps...

Simplification

Comes in roughly two forms: geometric and attribute.

In both cases, reduce the information content and resolution.

Exaggeration

Oh, yes it happens. But somehow this is more an issue of a collision between other components of the mapping process, not an objective in itself.

Example: river, railroad tracks, road. The road usually cannot be shown as close to the river as it is in fact, because the railroad symbol is wider than the actual railroad (at scale)....

Symbolization

Finally, we get to the heart of cartography: Using symbols to play the role of objects.

Mentions imagery (false color and photographic process - see lectures on remote sensing and images (06,07).

Core of symbolism: graphic semiology (systems of signs)

• SIZE
• SHAPE
• PATTERN
  • texture
  • spacing
  • orientation
• COLOR
  • Hue
  • Darkness (Lightness, "Value")
  • Saturation (Chroma)

(other cartographers would alter this list slightly, but it will do for us)

In Map Library exercise, you will be looking for class intervals and map symbols...

Version of 9 December 1999