Week 6: Following Rivers Through Time

top of page

Getting to Know Classification in ArcGIS

So far the information displayed on the Louisiana GIS map gives you a good overview of where the Cities and Parishes in Louisiana are located. However, in order to communicate more in-depth information about the map's features we are going to modify the map symbols to not only show location, but also to display additional information about the data they represent, such as numbers of homes or population. This type of representation of data is known as classification.

Launch ArcGIS, Open the Louisiana Project File, and Add a New Layer

• Launch ArcGIS by double-clicking its icon on your desktop or by clicking its icon in the or Launch Bar.
• Choose File > Open, navigate to c:/eyesinthesky2/weeks/LouisanaAM, select the LA_Rivers.mxd file, and click Open.
  
• A map of Louisiana is displayed when the project opens. Parishes are shown in tan and the Mississippi River is blue.
• Click the Add Data button, navigate to the LouisianaAM folder, select cities.shp, and click OK. Cities should now be the top layer on your map. Note: The cities layer is automatically turned on when it is added.
Show me what this looks like
Hide

top of page

Classify Point Features by Field

Open the Layer Properties window for the cities layer by right-clicking the cities layer in the Table of Contents.

Choose Properties from the context menu. Then move the Layer Properties window so you can see both it and the map at the same time.

top of page

Classify Point features by Equal Interval

In the Properties window for the cities layer, graduate the color of the symbols to emphasize the Populationof Louisiana cities in the year 1990. Select the following options:

• Draw features using: Quantities then Graduated Color
• Classify by the Field: POP_90
• Classes 5
• Style Circle
• Click ClassifyButton
• Classified by Equal Interval,click OK
• Color Ramp = Yellowto Red
• Click OK
Show me how
Hide

Right click on the layer name cities to open the Layer Properties window.


• Draw features using: Graduated Color
• Classify by the Field: POP_90
• Classes 5
• Style Circle
• Classified by Equal Interval
• Color Ramp = Yellowto Red
• Click OK


This is the completed map. The resulting map consists of mostly yellow dots. In fact, 386 of the 391 cities are represented by small yellow dots.

Thought question:
Study the map - How do you think equal interval classification breaks up the data?

The map now displays the location and population of each city in Louisiana. The data is differentiated both by size and color. It is split into 5 groups with the cities that have the largest population colored with red dots. However, the data on the map is hard to interpret. Notice that all the cities are almost all of one size. Return to the Properties window and adjust your settings.

top of page

Classify Point features by Quantile

There are other ways to break data sets into groups. Try separating the population data into quantiles. In the Properties window, select the following options:

• Draw features using: Graduated Color
• Classify by the Field: POP_90
• Classes 5
• Style Circle
• Classified by Quantile
• Color Start color Light Gray, End color Red
• Color Ramp = Yellowto Red
• Click OK

Thought question:
How do you think Quantile classification breaks up the data?

top of page

Compare the difference in these two methods of classification

• Return to Layer Properties window.
• Under Classified by, switch between Equal Interval and Quantiles. Examine the Range and the number of Records in the Break Values window.
Show me the difference
Hide
These are the two Properties windows. Look at the Range of the data and the number of Records.

Equal interval classification breaks up the population data into groups having an equal range of values (i.e. 0-10, 11-20, 21-30, etc.). In this case, when Louisiana cities are mapped by equal interval, it highlights the fact that most of the cities have low population.

Quantile classification on the other hand, breaks up the data into groups having the same number of features (i.e. 10 per group, 50 per group, etc.). However, this type of classification can be misleading.

top of page

Classify Point features by Natural Breaks

In these first two types of classification, the data is not represented in a way that helps reveal patterns on the map. Why, you might wonder? Carry out a query to do a little more investigation of the data. Set up a query that asks how many cities in Louisiana have a population greater than 220,000.

Make queries for these other questions:
How many cities have a population between 220,000 and 100,000?
How many cities have a population between 100,000 and 40,000?
How many cities have a population between 40,000 and 10,000?
How many cites have a population less than 10,000?

Take another look at the Attribute table for the data and observe the patterns in the data. Before setting up a third type of classification, Natural Breaks. Note: We are going to do this manually.

To set Manual breaks - In the Classification window choose Manual, in the Breaks Values window, select the first break, 10000. In the next break type 40000 to reflect the next break change the next level to 100000. Repeat this procedure for all of the five breaks listed below.

(Disclaimer: this can be a bit hard to accomplish, so if you are finding this difficult, just skip this step and look at the pictures, below.)

• Less than 10000
• 10,000 - 40,000
• 40,000 - 100,000
• 100,000 - 220,000
• More than 220,000

Click OK and close the Layer Properties window.

top of page

Classify Polygon Features by Field

top of page

Classify polygon features by unique value

It is also possible to classify polygons in order to display a variety of information. Previously, you investigated the data in the attribute table. Classification allows you to select and display that information on the map.

• Turn on the Parishes layer.
• Classify all the Parishes by Categories > Unique Values, in this case by NAME.
• Choose Field NAME. Click on Add All Values.
• Color Scheme: Random, Style: Solid fill.
• Click Apply.
• Before clicking OK, explore other color options for the fill for each Parish.
  Show me
  Hide
  results
  
  
  
  

While still in the Properties window, click on symbol next to the value you are symbolizing, this will bring up a color picker window. From the Swatches window you can change the colors of any of the fill values.

Another method to choose a color palette is to use the Color Ramp pull down menu and choose a different palette of colors, such as minerals or pastels. If you want to see the word descriptions of the Color Ramps, right click on the color bar and uncheck the Graphic View.Explore these other options on the map.

When working with multiple layers, the Transparent color fill can be useful. This allows the map to show boundaries of a polygon, as well as the data layer below it on the map. For example, if you had an image layer such as a NEO land surface temperature image and you wanted to overlay political boundaries over that image, this would be the technique that you might use.

• Change the Parishes fill to transparent. In the Symbology tab, choose Features, Single Symbol.
• Click on Symbol choose Fill Color: No Color fill.
• This choice leaves only the outline on the polygons.
  This technique is very handy when you want to be able to see through a reference layer to a layer below.
Show me how
Hide
Changing the fill layer to transparent.

Parishes is transparent.

top of page

Classify polygon features by quantile

Just like in point layers, it is possible to display multiple types of information on the map by classifying the polygon layers. To illustrate this idea, change the classification of the Parishes layer to display information about Population density.

• Right click on the Parishes layer and open the Properties window.
• Select the following options:
• Draw the features using: Quantilies > Graduated Symbols
• Fields Value: Pop90_SQMI (population per square mile in the year 1990).
• Choose Classify by Quantile, Click OK.
• Click Apply and click OK.

The Parishes are now color coded by Population Density.

Observe the relationship between the cities layer and the Louisiana Parishes layer. It comes as no surprise that the Parishes with the largest cities also have the highest population density.

• Turn off the cities and the Louisiana Parishes layers

top of page

Add Historic Settlement Patterns and Rivers to Discover the Relationship between People and Rivers

Up to this point, you have captured one year's worth information on the map and with that year's data told a story about the population of Louisiana in 1990. However, maps can tell stories that explore both spatial and temporal questions.

Now let's explore the question, "How was the State of Louisiana populated over time?"

• Add the historicsettlement shapefile.
• Add a new layer to show the historic settlements in Louisiana. Click the Add Data button. Navigate to the Louisiana folder. Click once on historicsettlement.shp file to select it. Then click Add.
• historicsettlement should now be the top layer on your map. Notice that the layer is automatically turned on when it is added.
Show me
Hide

1. Click the Add Data button.
2. Navigate to the Louisiana folder.Click once on historicsettlement.shp to select it. Then click Add.
3. If necessary, move the new layer down in the layer list. Put it below the point and line layers. Notice that when it is added it is all one color, not very informative!