Introduction to Geographic Information Systems in Forest Resources

UW Home GIS@UW Search

Syllabus Schedule Class Meetings Assignments Course Data Internet Search Current Grades Contact Us CFR 590 Internet-only section Lab Locations

Discussion

Cartographic choices are the basis of using maps for communication. How you decide to represent your data will determine what the map communicates. You can draw the reader's attention away from one detail and to another. You can distort data so that the reader gets a skewed picture of the underlying reality of the map data.

Consider these 2 maps; both show the population of each state, where the darker color represents higher population. The first map shows the total population, while the second map shows population density (population per unit area). In the first map, California shows the largest population, and most of the Northeastern states show low population. However, in the second map California shows somewhat low population density, while the Northeast shows the highest population density. Both maps represent the same general thematic data (population), but each uses a different cartographic method, and each map communicates a different idea.

 

Frequently, different map symbols are chosen because of their psychological impact (for example, green and blue solid fills are often used to show "cool," or non-threatening areas, while red is used to show "hot," or dangerous areas). Line weights also impart relative importance to different linear features (interstate highways are usually drawn thicker than county roads). Different point symbols can be used to signify characteristics of places (National Park maps often show amenities such as boat launches, trailheads, picnic areas, and restrooms with specialized marker symbols). Think carefully about what you want to communicate, and choose your map symbols accordingly.

Classification is another method of changing what you want to communicate with your maps. Research has shown that it is more difficult to comprehend or compare a large number of objects or values. Classification allows you to lump similar features into classes; fewer classes are easier to compare to one another. In the maps above, if any more classes were used, the classes would become visually indistinguishable from one another.

Any good GIS software will give you the ability to quickly and easily change the way maps are displayed, by changing symbols, colors, or classification schemes

Thematic mapping with the Legend Editor
Choosing a legend type
Classifying map displays
Normalizing data
Manipulating classes
Modifying legend elements
Sorting values & labels
Flipping symbols
Ramping colors
Undo changes
Working with null values
Changing symbols with the Symbol Window
Scaling symbols
Rotating symbols
Offsetting lines

Thematic mapping with the Legend Editor

The ArcView Legend Editor allows you to change the symbology of theme features within views. There is a great deal of flexibility about how to display features. Many built-in symbol palettes are available for area, line, or point features, and it is also possible to create custom symbols.

To change the symbology for a theme, make sure it is active. Then do one of the following:

• Click the Edit Legend button ,
• Select Theme > Edit Legend from the menu, or
• Double-click on the theme name in the Table of Contents.

All of the cartographic changes described in this section are accessed via the legend editor.

Several legend types are available:

• Single symbol (all features of a theme will be drawn with the same symbol)
  Use a single symbol when you only need to display features' locations.
  
  
  
  
• Unique value (features with common attribute values will be drawn with the same symbol)
  Use this to display categorical data, such as soil type, county, or eco-regions. Be careful with unique values display, as a large number of unique class values can easily make classes indistinguishable.
  
  
  
  
• Graduated color, also known as choropleth (colors will gradually change with changes in a numeric attribute value)
  Use graduated color to show data that progress over a range of numeric values, such as stand age, temperature, or precipitation. This map shows the establishment year of forest stands. Note here that polygons without establishment years are not displayed.
  
  
  
  
• Graduated symbol (size of point markers will increase proportional to a numeric attribute value)
  Graduated symbol displays are useful in displaying the relative magnitude of numerical attributes for point data. In this example, forest stand centroids are being displayed with dots that are proportional in size to site index (a proxy measure of forest productivity).
  
  
  
  
• Dot density (a numeric attribute value is used to assign a number of dots representing relative frequency)
  Use dot density to display how an attribute (such as number of elk or trees per acre) is distributed over a large area. Note: the dots are placed randomly and do not represent the true location of features!
  
  
  
  
• Chart symbol (a bar or pie chart is used to represent the value of several different attribute fields at once)
  Chart symbols are good for comparing a number of attributes at the same time. Here, we can see a chart that displays the size class, stocking level, and site class for each stand.
  
  
  
  

Classifications are used to group features in similar classes. Classification methods can only be used for numeric attributes.

Classification methods (type of classification, number of classes) are only used when displaying data using either graduated colors or graduated symbols.

The 5 classification methods ArcView uses are:

• Natural breaks
  ArcView attempts to find clusters or concentrations of data and place class breaks between the clusters. By default, the number of classes is 5, although you can change the number of classes if you wish. Here are stands displayed in 5 natural breaks classes, according to age.
  
  
  
  The chart shows the number of acres in each unique age class. ArcView uses an algorithm to separate classes based on the spread of the data (in this case, the spread of age).
  
  
  
  
• Quantile
  Quantile classification creates classes of roughly the same number of individuals. In this example, in order to get quantile classes, younger stands are grouped with older stands. Each class here will have approximately the same number of forest stands.
  
  
  
  
• Equal Interval
  Equal interval classification divides the range of values into intervals of roughly the same size, and then displays features according to the class they fall in. Here, are equal interval classes of 39 years. Because the range of ages is large, but most stands are young, the oldest stands fall in a class by themselves.
  
  
  
  
• Equal Area
  Equal area classification divides a group of features into classes which are roughly equal in area. Equal area classifications could be used in determining areas for allocation of scarce resources. For landscapes with similarly sized features, equal area classification will appear similar to quantile classification. Here, each age class will contain polygons whose summed area will be approximately equal. For example, the 68-year age class contains roughly the same area as combined stands ranging in age from 71 to 198.
  
  
  
  
  
• Standard Deviation
  Standard deviation classification will show the variation in data around the mean value. This shows the relative spread of attribute values. Here, the oldest stands are far outside the mean stand age, and show up the darkest red.
  
  
  
  

Normalizing data

Normalization of data values expresses attributes as relative values rather than absolute.

Normalization can express an attribute as a percentage of the total. For example, a map could be made to show each stand's contribution of fiber in comparison to the total amount of fiber on the forest.

Alternatively, normalization can express the ratio of one attribute to another. For example, a map could be made to display the per-area contribution of fiber to the forest total. The best example of this is the raw population vs. population density maps from the top of the page, in which the population field is divided by the area field.

Do not normalize attributes that are already expressed in percentages or densities (these data are already normalized).

Manipulating classes

It is possible to alter the classification by adding and deleting classes, by altering the position of class breaks, or by editing class labels. Here are examples of both manipulations. Both of these maps display the same data, but with different numbers of classes, different colors, and different class labels. When do you think one of these would be used rather than the other?

The Legend Editor is used to make changes to legend elements. It is used to alter the legend type, number of classes, colors, normalization, and labeling.

Adding and deleting classes

Classes can be added to and deleted from the legend using the Add Class and Delete Class buttons. If you want to increase or decrease the number of classes, use these buttons.

Sorting values & labels

Legends can be sorted in ascending or descending order. The Value and Count fields are sorted numerically, and the Label field is sorted alphabetically. The Sort Ascending or Sort Descending buttons are used for sorting.

Flipping symbols

The order of symbols can be reversed by clicking the Flip Symbols button . The order of the symbols will change, but the order of the values will remain the same. For example, if your colors ramp from green to yellow to red, flipping the symbols will reverse the order of the colors.

In this example, the red monochromatic 5-class legend is flipped. Note that the values are the same, but the colors have changed.

Ramping colors

Color ramping is automatic when you choose the graduated color legend type. However, if you are want to change the first and last color, and have them gradually change between, use the Color Ramp button . You can also select one or more colors in the middle, and ramp the colors that fall between (e.g., blue > green > yellow > red > white).

In this image, the color ramp goes from green through yellow to red.

Random colors & symbols

Random colors and symbols can be selected using the Random Colors and Random Symbols buttons. Use these in combination like so only if you want to nauseate the viewer:

Undo changes

If you are not satisfied with the legend changes you have made, the Undo Changes button will return to the previous legend that was applied (up to 5 previous changes). Any changes before the previous 5 will be lost. There is also no method for redoing an undo.

Working with null values

Sometimes attribute data contain null values (often coded as 0 or -9, numerically, or blanks in character fields). To remove these from the display, press the Null Values button . This will allow you to enter the values that will not be displayed. When you specify the null value, any features with this value will not be displayed.

Changing symbols with the Symbol Window

Changing individual symbols is performed with the Symbol Window. The Symbol Window contains 5 different palettes, each controlling a different symbol class. The palette windows control

• fills (area pattern)
  
  
• pens (line width and line pattern)
  
  
• point markers (symbol and size)
  
  
• fonts (typeface, style, and size)
  
  
• colors (color of the above symbols or text)

The Symbol Window also contains a Palette Manger, which allows you to save and load custom palettes and icons.

Scaling symbols

Marker symbols can be scaled so that as you zoom in and out, the size of the marker changes with the scale. Without scaling, as you zoom in, the absolute size of the markers will stay the same; with scaling on, the markers will increase in size.

Here is a view with simple point symbols:

This is a larger scale view with scaling off:

And with scaling on:

Rotating symbols

If a field exists within a point theme's attribute table describing the angle of the feature, this can be used to rotate each marker by that value. Marker symbol rotation can be used to show the direction of travel of a point phenomenon (e.g., an animal's direction of migration based on a telemetry point or wind direction from a weather station).

Offsetting lines

Lines can be offset any distance you like. This is handy when you are displaying several lines that lie on top of each other (for example, the Pack Forest boundary is delineated in places by roads and streams). Offsetting allows your map to display all lines. The offset direction is controlled by the arc's direction.

Here are similar views, one with line offset at a value of 0 (no offset), and the other at a value of 2.

Syllabus Schedule Class Meetings Assignments Course Data Internet Search Current Grades Contact Us CFR 590 Internet-only section Lab Locations

The University of Washington Spatial Technology, GIS, and Remote Sensing Page is provided by the College of Forest Resources and the College of Ocean and Fisheries Sciences through Unit-Specific UIF. Site administrator: Phil Hurvitz.

Copyright © Phil Hurvitz, 1998-2003