Cartographic Skills Module 7: Choropleth Mapping

Module 7 lab continued the data classification experience started in Module 6 and involved the completion of two choropleth maps.  Both ArcMap and Adobe Illustrator were used to complete the maps, one of which was in color (Change in US Population by State) while the other was in gray scale (Change in US Population by Census Division).  The exercise included using Excel for making calculations and utilizing that information to determine appropriate classification method for the data, selecting appropriate color scheme and grays for the different classes of the choropleth maps, labeling states and divisions, and creating legends.  Being quite fond of Excel for numerous purposes both personally and professionally, I used it beyond the lab requirements to determine additional information for different classification methods which I then used to select a data classification method for the second map.

Percent Change in State Populations (1990 to 2000)

Practicing projections in Introduction to GIS (GIS4043) provided the experience needed to change the projections of Alaska and Hawaii to something which would yield more realistic graphics, specifically for Hawaii.  The nearly North-South orientation of the Aloha State was a bit disconcerting, but reprojection took care of that.

Labeling divisions for the second map was an exercise in experimenting effort.  Several font sizes, types, and effects were attempted in trying to achieve the best, most uniform appearance for the division labels.  The appearance of masking was not appealing, and trying to label enough of the states in each division to distinguish between adjacent divisions which were in the same class (West North Central/East North Central and Middle Atlantic/New England) was a challenge.  Finally I settled on allowing labels to straddle some state lines, but shifted the labels so that the lines were generally between letters or words in the label.  Using Transform > Move allowed very fine movement to do this.

Percent Change in Census Division Populations (1990 to 2000)

The results of the lab assignment are two easy-to-read-and-interpret maps, one for the change in US population for individual states as well as the District of Columbia and another for the change in US population by Census Divisions.

Cartographic Skills Module 6: Data Classification

Data classification was the focus of this week's Cartographic Skills Lab.  The creation of a map comparing four different methods of data classification (Equal Interval, Quantile, Standard Deviation, and Natural Breaks) was completed by creating four data frames within one map and then applying a method of data classification to each of the data frames.  As careful as I thought I was being, I learned the hard way that an oversight can produce an erroneous map. Printing out a review copy of the map prior to submitting it helped me to catch a couple of major errors.  Nice-looking maps aren't especially helpful if the information is wrong.

Comparison of Four Methods of Data Classification as Applied to
Black/African-American Percentage of Population
Escambia County, Florida, 2000

After correcting the errors in the data classification method comparison map (involving legends for certain classification methods being applied to the wrong data frames), it was easy enough to save the map as a new document and then make some adjustments to complete a map of the data classification method which I thought best for this particular purpose.  In this situation I felt the Natural Breaks method was best suited.  The Natural Breaks method takes into consideration clusters of information which can be overlooked by using the Equal Interval method.  The Standard Deviation method doesn't seem to be very user friendly for this purpose.  The Natural Breaks method puts more emphasis on groups of similar data and is easy to interpret which seemed appropriate for the audience likely to use a map like this.

Natural Breaks Method of Data Classification Applied to
Black/African-American Percentage of Population
Escambia County, Florida, 2000

GIS4043 Projections Part 2/Data Search

As would be expected, the Projections Part 2 Lab continued to build on the skills learned in the previous lab.  The practice exercises preparing for the actual lab project were entertaining, although I will admit to being a bit disappointed that the bald eagle nest data was for Santa Rosa County instead of Escambia County.  At some point I had accidentally deleted the coordinates in my handheld GPS unit for the bald eagle nest on the UWF campus.  I know that it's in the vicinity of the cross country trails, but with the vegetative growth over the past few years, I can't spot it any more.  At least I now know where to locate a source of data for that information!

The final portion of the lab involved the practice of obtaining, defining and projecting datasets. After aerials were loaded into ArcMap and checked against a basemap, vector files were reprojected and added to help identify the subject area.  Finally, an Excel document containing data regarding the petroleum storage tank contamination monitoring sites was edited to convert coordinates in DMS format to DD format to keep ArcGIS satisfied.  (Surveying and geocaching provided plenty of experience for that exercise.)  After all the information was loaded into ArcMap, prettying up the map was all that needed to be done.  I decided to run some queries to sort out tanks by current status.  I'm happy to report that in the area covered by the Perdido Bay and Fort Barrancas quads, there are no abandoned tanks.

Petroleum Storage Tank Contamination Monitoring Sites
Perdido Bay and Fort Barrancas Quadrangles, Escambia County, Florida
Coordinate System: NAD 1983 State Plane Florida North FIPS 0903 (US Feet)

Defining and projecting datasets throughout this lab helped me realize the extensive versatility and applicability of GIS to so many situations.  Many potential uses have come to mind already.

  

Cartographic Skills Module 5: Spatial Statistics

The map below was created while progressing through ESRI's online course, "Exploring Spatial Patterns in Your Data".  This exercise involved several tools from the Spatial Statistics toolbox in ArcMap.  After weeding through some cobwebs to recall my statistics education, I found this to be a fascinating way to visually show the mean center, median center, and directional distribution of the data.  Having this visual representation of the data is helpful in determining whether there are potential issues with the data set which can also be confirmed with additional analysis using some of the other tools in the Spatial Statistics toolbox.

Analysis of Data: Weather Stations in Western and Central Europe

While the creation of this map, the histogram, and the QQ plot were rather straightforward, I found the latter part of the online course to be far more foreign.  Additional work with Voronoi maps, semivariogram clouds, trend analysis, and the like will be necessary before I feel confident working with them.

GIS4043 Projections Part I

The first part of this week's lab focused on comparing three different projections of Florida and how they affect area.  Flipping back and forth between the Florida projections in the ArcMap was helpful to see some of the differences inherent in projections. Until this lab I had no idea that there were so many projections in use.  The more I use ArcMap, the more I learn that there usually are multiple ways to complete any one task which has been quite helpful.

This particular map did not include an inset map.  Since the purpose of the map is to compare differences in area caused by different map projections, I figured that it would be used by someone either familiar with the location of Florida already or who was primarily interested in the concept of comparing projections themselves in which case the region to which the projections are applied is of lesser importance than the comparison itself.  A locator map would be extraneous in that case.

During the process of creating a map, ideas for improvements or different aspects to explore usually lead to changes in the original plan.  So even though I'll sketch a preliminary map, I may end up with a totally different product for submittal, sometimes having to choose from three or four different workups that evolved out of the initial sketch.  In some respects this is no different than working up different product options for a client to preview before committing to a final.

With this particular map, viewers can see that different projections can affect area figures, and so if accurate area is a priority, then a suitable choice of projection is imperative.  I'm looking forward to developing the expertise necessary to do that.

Four counties across Florida are compared for differences
in area calculated in three different projections.

GIS3015 Module 4 Lab: Typography

This lab focused on utilizing Adobe Illustrator to create a map of the Middle Keys of Florida. The emphasis of this exercise was on typography.  The importance of layer organization cannot be stressed enough!  I found myself re-arranging and prioritizing my layers, making adjustments as I discovered issues here and there.  Over time, I anticipate that this will become even more logical and easier to complete. Being new to AI, I experienced some difficulties caused by inadvertent key strokes and who knows what else. These were all resolved, except for being unable to draw a straight-line path on which I could place text.  No online source of information was located that could help me eliminate the error message which kept popping up, so that will require further investigation if it happens again in a different document. Practicing letter and word spacing, kerning and leading during this assignment helped me to understand the advantages of those practices better.  Creating a legend in ArcMap is preferable to creating one in AI, in my opinion.  By the end of it all, a map of the Middle Keys of Florida showing cities and select features was produced.

Map of Middle Keys of Florida showing cities and other features

GIS4043 ArcGIS Online and Map Packages

This lab focused on developing and submitting map packages to ArcGIS online. It has been the most time-consuming lab to date for me.  One map package featured climbing areas and points of interest in Yosemite Valley, while the other focused on the study of trees in the Aguirres Springs drainage.  These procedures are valuable, but I can see that much more practice is needed, so that far less time is spent on referencing multiple sources of help to determine exactly what is needed and how that can be accomplished.  This lab also reinforced that knowing exactly what outcome, result or product is expected must be very clear.  If the goal is not readily determined, prompt clarification must be requested.  This would be of benefit to the producer as well as the client.  The ability to use map packages will be very beneficial, and I'm looking forward to the time when completing similar tasks will be second-nature.

Submittal:  Overview of Map and Tile Packages

Submittal:  Optimize a Map Package

Module 3 Lab: Cartographic Design

The purpose of this week's lab was to continue learning Adobe Illustrator to develop cartographic design skills during the process of creating a map of the Hispanic populations of south Florida counties.  The cartographic process was reinforced during this exercise, including the cartographic design aspects of visual weight, contrast, figure-ground, and balance.   This particular assignment was more involved than others, requiring a lot of rearranging of the choropleth map and essential elements, trying different paper orientations, printing out samples, and critiquing to achieve a satisfactory final product.  Throughout the exercise many of my AI skills got quite a workout and many new skills were learned, with just a few gasps and quick touches of Ctrl Z to undo scary errors.

Census data provided by UWF was already grouped for the legend, and those figures were adjusted per lab instructions.  A greater-than sign (>) at the beginning of each percentage class after the initial one indicates a clear distinction between adjacent percentage classes. Curiously, Charlotte County is missing from the map.  Perhaps this is due to a Hispanic population less than 3.3% or due to missing information in the data file.  Further investigation is warranted.  For clarity and to use up some white space, I added labels for the Atlantic Ocean and Gulf of Mexico.

Module 3 Lab:  Hispanic Populations of South Florida Counties

After numerous revisions and four potential submittals, I believe I have a map which provides an easy-to-read and easy-to-interpret collection of information regarding the percentages of Hispanics living in south Florida counties.