I initially conceived this map while trying different things with world data but decided to add a different twist to the manner in which the data was displayed. Most people choose to display this type of data as “persons per sq km,” and I decided to do the exact opposite, to see how much space there is for people of each country.
The data for this map were collected from that which accompanied the software for ArcGIS, by Environmental Systems Research Institute, Inc. (ESRI). It is a simple join between population and country data, and I found it to be a rather interesting display, though I usually do not care for these types of symbols on maps. They always seem too difficult to ascertain exactly what is going on in the map, unless there is a huge disparity.
I encountered a problem right away, when I chose the proportional symbol. All of the symbols were so big that they covered most of every country. I quickly figured out that I could scale down the size of the initial symbol, so that the rest were not so obtuse, and was quickly on my way to creating the rest of the map. The only thing that still concerns me is the relative size of the lower numerically represented symbols.
The color that I chose to represent the base was a little more complicated, and I ended up using two separate world layers to get the effect that I wanted. I thought that the continents should be green, symbolizing mostly habitable areas on the planet, while at the same time, I expected Antarctica to show up as white, since there is usually snow there all the time; thus, the two layers, one white, the other green. It was at this point, that I chose the symbol color. Yellow seemed to be the obvious choice. It typically means happiness, and I figured that the more space there was for each person, the happier they would be, so it was good fit.
The background gave me a small problem. While it was easy enough to turn the oceans blue, it also turned the entire background of the map blue. There was no way to distinguish where the earth left off, and the canvas began. Initially, I tried adding a graticule. After reviewing it, there was some sense to having it, to understand the distortion in the projection that was taking place. However, it really did nothing to offset the earth. Ultimately, after the map was completed in ArcGIS, I imported it to Photoshop and dissolved the background around the globe. I changed the background to a light gray and left it at that.
The projection was a little difficult to come to terms with as well. I tried several, but they distorted area too much, or a few of symbols completely obliterated the areas they represented. In a final act of desperation, I turned to the Cartography book and discovered my answer there. It indicated that the Winkel III World Projection was a good one to use, so I went with that. It still did not fit the situation though. Countries were still being blocked out by symbols. However, the Winkel II worked perfectly to preserve area and distance, while preventing countries from disappearing under symbols too much. Greenland, Svalbard, and the Falkland Islands still have this problem.
Another problem I encountered was label placement. Ultimately, I raised the symbol weight on the features to a point that they finally began to show only the most prominent relevant features. In addition, since there were really no oceans in the shapefile, I had to add the labeling for those independently. In addition, since the countries previously mentioned were obliterated by the symbols, I went through and manually added labels on top of those symbols, for easier recognition.
At this point, the remainder of the map was rather straightforward and simplistic. A self-explanatory title was added, along with the data credits, as well as the legend. As with most legends, I declined to label it as a “Legend;" it just seems silly, unless there is so much activity on the map page that you cannot discern the legend from the map. Of course, in that case, it might be better to start over with a new map.
Finally, the final product was created ready for press, in 300 dpi JPG image format. The results were rather surprising though. I did not expect to see such a difference in the US and many of the other countries. However, I believe that this data to be heavily weighted towards the three obvious places with the most area per person. If those were removed, the map might show a completely different picture.
Overall, much like graduated symbols, I still do not care for these types of maps. I suppose that it might prove useful for displaying some data, but this is probably not the best format for its display. It does show some recognizable trends though, that would be more apparently in graduated color map.
GGP330 - Cartography
No comments:
Post a Comment