Sadie Trush
10/5/2015
Lab 1 Questions
Q1: The Source tab provides: Data Type: (Shapefile Feature Class), Shapefile (R:\Geog482_2\Student_Data\trush\Lab_1\Data\cb_2014_us_state_500k.shp), Geometry Type (Polygon), Coordinates have Z values (Yes), Coordinates have measures (Yes), Geographic Coordinate System (GCS_North_American_1983), Datum (D_North_American_1983), Prime Meridian (Greenwich), and Angular Unit (Degree).
Q2: This layer coordinate system is GCS_North_American_1983.
Q3: The Albers Equal Area Conic projection shows the shape of the continental US in the way I am most used to seeing it and would call “proportional” or “normal.” The Mercator projection is straighter across the top of the continental US, and appears a little blockier to me. The Carree projection stretches the States from what I am used to seeing. The Robinson projection makes the map appear almost slanted.
Q4: The position of cities in relation to each other doesn’t appear to change greatly between the Albers Equal and Mercator projections. It may increase slightly between Seattle and Chicago since the Albers Equal is not as linear as the Mercator. Distance between New Orleans and Boston may also increase slightly. The Carree Projection greatly increases distance between major cities on the West and East coasts (compared to the Mercator and Albers Equal). The Robinson projection also increases distance between West and East coast cities, particularly Los Angeles and Boston.
Q5: Albers Equal Area Conic distorts the shape and linear scale (but this is minimized with the use of two standard parallels in a conic projection. It is best when looking east to west. The Mercator projection distorts by stretching vertically (especially towards the poles – it will make land area further away from the Equator appear larger). When using the Plate Carree projection, almost all distortions increase with distance from standard parallels. Distance is correct along standard parallels and meridians as well as general direction, but shape and area are more greatly distorted with distance from the standard parallels. In the Robinson projection distance and direction are the most distorted, while the distortion of shape and rea are low within local parameters. It is often referred to as a “compromise map” (arcGis online).
Q6: The planar distance using the Robinson projection between LA and Boston in the 4,572, 721 meters, and 1,485,306 meters between LA and Seattle. The planar distance using the Plate Carree projection between LA and Boston is 5,343,184 meters, and 1,570,338 meters between LA and Seattle. The planar distance using the Mercator projection between LA and Boston is 5,391,124 meters, and 2,044,913 meters between LA and Seattle. Using the Albers Equal projection, the distance between LA and Boston is 4,156,735 meters, and 1,547,789 meters between LA and Seattle. From this, we can see the Mercator projection stretches the map horizontally the most, and the Albers Equal stretches it horizontally the least. The Robinson projection has the least vertical stretching, while the Mercator projection does the most vertical stretching.
Q7: This dataset contains variables of Shape, StateFP, Statens, AFFGEOID, GEOID, STUSPS, Name, LSAD, ALAND, AWATER, FID GEPD10, STUSPS10, NAME10, ALAND10, AWATER 10, INTPTLAT10, INTPLTLOM10, and DP00 numbers that code for characteristics of the populations such as ‘population over 85 years’ (that I chose), or population female, population male, population under 5 years of age, etc.
Q8: I chose to map population of the CONUS over 85 years old. I used a neutral blue color scheme to avoid any connotations associated with other colors (like red), and I there are no bodies of water or oceans that concerned me in this map. The preperscribed 4 classes given automatically by ArcMap made my map seem too skewed towards one end of the color spectrum or the other. So, I chose to create 8 classes so a user can observe the gradual color change and states with “inbetween” populations of people over 85. For my first map I used Jenks Natural Breaks classification, which identifies the natural grouping of data. While this is a good form of classification because it allows you to see the natural trends of a data set, it also has a bias to exaggerate the difference between groups. This can make groups seem more defined than they actually are. For my second map, I used Equal Interval classification. This style of classification breaks the data range equally into a specified number of intervals (in this case, 8). It’s bias is that it can force certain states into higher or lower classes even if they are very close to the higher or lower end of that class interval. In my map, this resulted in most the States being lighter in color, and less than half a dozen showing heavy concentrations of people over 85. My last map used Quantile classification. This form of classification puts an equal number of data pieces into the set number of intervals. The bias of this method can be great particularly if you are not working with linear data. It forces data into intervals to represent them equally by color scheme. This was evident on my map.