(A very brief) History of Cartography

Prehistoric maps are preliterate: bone, skins, cave paintings to show locations of things, routes, etc.

• Most of what’s available and written up is western cartography; it’s believed that mapping has gone on longer than the evidence we have (12^th C), but without evidence…
  
  
• Oldest recorded route/plan: 9ft wall painting, found in 1963 of a town plan, showing buildings and a volcano. Found in Anatolia, dates 6100-6300BC.
  
  
• Oldest surviving maps are Babylonian clay tablets (approx 2300 BC) and Egyptian drawings. Egyptians were strong surveyors (Nile flooding, for example). First maps dealt only with immediate area – “world” maps didn’t show up until later.
  
  
• Greeks did a lot to develop understanding of geography. By middle of 4^th Century BC there was a general acceptance by Greek scholars and scientists that the earth was spherical. Ptolemy (90-168AD) was a tremendous influence on western geography and cartography. Greeks did a lot to study size of the earth and its habitable areas, climatic zones and country positions. Also:
• Studied map production and globe construction
• Created gazetteer (with lat/long)
• Published “Geographica” in AD150, with world map, six regional maps, a few others
• Foundations of mathematical geography
• Produced a map of the known world (that stuck for a long time)
  
  
• Romans focused on practical uses: military and administrative needs. Few Roman (and Greek) maps survive, Ptolemy is considered “father” of many cartographic/geographic practices.
  
  
• Medieval maps restricted by church, little development during this period. Manuscript maps in England.
  
  
• Islamic (al-Idrisi, who now has a software named after him), made world map (1154). Compass and 360 degree system widely in use.
  
  
• Chinese mapmakers – oldest known map is 1137. Independent development. See History of Cartography Project.
  
  
• 1200-1700s: new discoveries and exploration lead to new mapping, mapmaking as art and science. Better instrumentation led to better (more accurate, geographically) maps. Mercator makes new map of Europe (1569), becomes common world projection (used for navigation – more on this in subsequent lecture). Ornate maps created, printing press allowed for better distribution. Many atlases produced in 1560s-1620s.
  
  
• Dutch were the world power in the 1600s, and the center of cartography: atlases, sea charts, town plans, etc. Excellent engraving and design practices. (Followed by London, France)
  
  
• Up to present time: Again, more developments in science led to better accuracy (Longitude, John Harrison, 1750s).
• National military, colonization needs
• World wars (aerial photographs, navigation charts)
• More mapping standards (international)
• Space technology (satellites)
• Computer mapping

So, what is a map…?

Types of maps and geographic data

Maps are only as good as the data that created them. The data are the building blocks, much like the statistics that go into a report. Some definitions of map:

• Any geographical image of the environment.
• A graphic representation of spatial relationships and spatial forms (Robinson)
• Mental maps (try this…)
• An abstract spatial representation of an environment.
• A representation on a flat surface of part of the earth's surface, to show physical, political or other features.

Maps can cover any area, big or small (neighborhood, city, state, country, continent, world, universe). In this class, we’ll focus on map reading and map interpretation (DEFINITION):

Goal: to learn how to communicate effectively with maps.

Every map has an author, a purpose, a projection … every map has a bias.

What about the data?

1. data acquisition
2. level of measurement
3. data inventory scheme
4. spatial prediction
5. derivation of values

1. Data Acquisition

Via:

• Direct perception (unaided human senses) – this is very subjective: we forget, we generalize, we make errors, etc.
• Ground survey measurements (horizontal and vertical, GPS units). More on this in projection lecture.
• Census, studying the nature of a population, aggregated by geographical unit. Tells us how many things are in a unit, but not exactly where (generalization). Bias inherent in units selected:
• Size. Large units may obscure crucial variation or patterns (eg, cancer rates). Equal-sized units in a region will work better for some areas than others. Detail is limited by unit size.
• Shape. Units should be fairly compact and of similar shape. But: compact as in shape, or compact as in homogenous?
• Orientation. Things like gerrymandering, for example.
• Always be aware of statistical methods used: how were outliers treated? What sort of variables were aggregated?
• Remote sensing (more in future lecture) – site-specific devices (eg, weather monitors) or environmental, surveying the environment.
• Compilation – using existing tables, graphs, texts, maps.
• Start with large-scale (more detail) and go to small-scale (less detailed) – errors are less likely to be greatly magnified.

2. Measurement levels

Two types: qualitative (what exists), quantitative (how many of these things exist, magnitude). Four levels, going from more generalized to less generalized:

1. Nominal. Most simple. Categorical (nom=name). Breaks data into classes. Data compared by type, eg, male or female.
2. Ordinal (ordered, or ranked on a continuum). No indication of magnitude, no numerical values: small, medium, large; safe, iffy, dangerous.
3. Interval. Numbers are used, but value is not absolute. Eg, temperature. 40 is more than 32, but 32 also equals zero. Also, calendar dates. 1^st, 2^nd, 3^rd…
4. Ratio. Absolute numbers, with a true zero (volume, length, etc). Best accuracy of the four. Has indication of magnitude. =, <, >, +, -, *, % etc (population density, income, distance).

3. Data inventory scheme

• Population counts, every member of a population. Can have high level of errors, due to instrument (interviewer), methodology (questionnaire), and human error (at all levels).
• Samples. (US Census does both of these.) Count a representative subset of your population. Eg, polls. Gives an estimate. Many statistical rules for selecting samples, for various uses, determining sample size, sample selection, methodology (random, systematic, stratified, etc.).

4. Spatial prediction (see textbook)

• Point to point, interpolation vs. extrapolation
• Area to point
• Area to area
• Point to area

5. Derivation of values

Statistical processing of raw data prior to mapping. Eg, ratios, indexes, regressions. Error can be introduced here, through poor selection of statistical method or bad application. Eg, deriving a mean:

Mean doesn’t show any trends in the data. Maps rarely show nature of data distributions used. Watch for things like average (income, temperature, soil type). Ask yourself, what was done? Why? What affect does this have on what I’m looking at?

---