Evans Week 3

Chapter 4: Mapping Density

• Why map density?
  • Mapping density shows distribution with uniform units
    • Larger tracts may have a larger number of people, but be more spread out
• Deciding what to map
  • Features: density of features such as number of houses
  • Feature values: density of values relating to features such as number of people per house
• Two ways of mapping density
  • By defined area: use if data is already summarized by area
    • Dot map
      • Each dot represents a specified number of features
      • Distributed evenly throughout area rather than clustering
      • Make sure that dots are large enough to convey information, but small enough to not obscure information
    • Shaded
      • Each area is shaded based on density, but it doesn’t show centers of density in large areas
    • Calculating a density value for defined areas
      • Add a new field to feature data table, assign density values by dividing the value by area of plot
  • By density surface: use if you have individual locations and need precise views
    • • Created in GIS as a raster layer – each cell gets density value
      • Provides the most detailed information, but takes the most effort
    • Creation
      • Cell size: units per cell, larger cells process faster but are less specific, smaller cells take more time but show a smoother surface
      • Display
        • Graduated colors: different shades of the same color, don’t use too many shades
        • Contours: lines such as that on a topography map, too few lines shows little detail, too many lines make it hard to read

I personally don’t like dot maps where the specific density is not shown and points are scattered across the area. I much prefer other types of density maps since, as a viewer, I am used to dots being used when there is a precise location shown and average values being shown with graduated colors. Dot maps that don’t show precise location are not as visibly appealing as graduated color maps showing the same thing, and I feel that they are more confusing to read.

Chapter 5: Finding What’s Inside

• Data features to consider:
  • How many areas?
    • Single area: monitor activity and summarize information such as service area around a facility
    • Multiple areas:
      • Contiguous: bordering each other
      • Disjunct: not connect to each other
      • Nested: area within an area
  • Are the features discrete or continuous?
    • Discrete: unique features
    • Continuous: seamless, must be summarized
  • Information needed from analysis
    • List, count, or summary
    • Completely or partially
  • Finding what’s inside
    • Drawing areas and features
      • Map shows boundary of area and features
      • Quick and easy, visual only, doesn’t give information about features inside
      • Making:
        • Discrete areas
          • Area shaded with light color and boundaries drawn on top emphasizes features
          • Area on top of boundaries emphasizes area
          • Outer boarder of area drawn with thick line, boundaries drawn in different colors shows discrete areas by category
        • Continuous features
          • Boundary drawn on top, details drawn in separate colors
    • Selecting the features inside an area
      • GIS selects a subset of features within the area
      • Tells information about features inside a single area, but doesn’t tell several areas as separate
      • Making:
        • You specify features and areas, GIS checks where each feature is and displays whether it is inside the area
        • Data table can be used to get information about features within specified area
    • Overlaying areas and features
      • Combines area and features into new layer or compares the two layers to get statistics
      • Displays information about features within several areas, but requires more effort
      • Making:
        • GIS tags each feature with a code for which area it is in, list of features or summary of attribute can be accessed
        • Attributes are stored permanently in the feature data table, so this can lend to deeper analysis

“Overlaying areas and features” maps seem the most interesting to me, but are clearly difficult to make as well. I hope I get to work on that during this course. I find interactive maps like that very cool, since they provide the most information in a shorter amount of time once set up.

Chapter 6: Finding What’s Nearby

• Defining Analysis
  • Defining near: is near measured by general area (100ft radius) or travel time?
    • Distance or cost: travel costs are costs such as money, time, and effort
  • Flat plane or using earth’s curves – smaller scale can operate on a flat plane, but larger needs curvature
  • How many cost ranges?
    • Inclusive rings: nested cost ranges
    • Distinct bands: 1-100ft, 100-200ft, 200-300ft ect
• Finding what’s nearby
  • Straight-line distance: within a circle around the area
    • Quick and easy, but gives approximation
    • Seems most helpful for less precise work like figuring out the general coverage of a store
  • Distance or cost over a network: actual travel times or cost along features like roads
    • Gives more precise cost over network, but requires accurate network layer
    • Most helpful for more precise work over roads or other transit lines, shows the actual time that it takes and can show traffic too
  • Cost over a surface: actual travel times or cost over an area not defined by lines like roads
    • Combines several layers to measure “open-world” travel cost, but requires data preparation
    • Most helpful for non-constrained travel such as walking in a forest rather than following side walks

This chapter is really cool since they’re not really maps all showing the same information in different ways like previous chapters, but showing really different information; they’re not really interchangeable because of the different uses. It’s also interesting how much information is needed to complete some of these; to create a cost over a surface map for travel in a forest, you would have to have data about the surface cover of the forest. To make a distance over a network map, you need an accurate network layer, and if accounting for traffic, you have to have information about traffic patterns in each place. While these are widely available to us now, it’s wild to think about how much information gathering has gone into that and is ready to be build upon.