IPM CRSP UGANDA

GIS WORKSHOP

USING ARCVIEW

Larry Grossman

Department of Geography - Virginia Tech


Held at Makerere University, Kampala, Uganda
February 19 - 21, 2002


VIRGINIA TECH
MAKERERE UNIVERSITY
NATIONAL AGRICULTURAL RESEARCH ORGANIZATION


NOTE: The materials on these pages--in printed and digital form and on the World Wide Web--are to be used only in association with the Virginia Tech IPM CRSP. Reproduction or use of these pages, in part or in full, for commercial purposes is strictly prohibited. Those wishing to reproduce or use these pages for non-profit, educational purposes should contact the author at LGROSSMN@VT.EDU for permission.


Acknowledgments: The digital data in these GIS lessons are supplied by the Institute for Environment and Natural Resources, Makerere University, Uganda; the National Environmental Management Agency (NEMA), Kampala, Uganda; the National Biomass Project, Department of Forestry, Kampala, Uganda; the World Resources Institute (African Data Sampler), New York; the Blackland Research and Extension Center and the Department of Rangeland Ecology & Management, Texas A & M University (Almanac Characterization Tool); and the United Nations.

Step 7: Classifying Data in Themes into Classes

We are now going to create maps showing the spatial patterns of agricultural variables.

Uncheck the boxes next to all theme names in the Table of Contents.

Add the theme "Uganda_maize.shp," which has data on maize production in counties and make the theme active. (Note: These data on maize production are not current and should not be used in your current research. They are used here for illustration purposes.) Display the new theme in our View and zoom to the active theme.

Note: I have deleted the themes "Uga_soil.shp," "Shambas.shp, and "Iganga-river_50000.shp" from the Table of Contents here, but you can leave them in your Table of Contents.

Right now, the theme "Uganda_maize.shp" is displayed uniformly throughout our map. We know where the various counties in "Uganda_maize.shp" are located--we can see their boundaries--but the map currently does not tell us anything else about the attributes associated with maize production in each county.

We can find out information about each county by using the Identify tool.

Examine the identity and various fields associated with a county.

Note that there are several fields about maize production, such as "Prod_ton" (total maize production in tons) and "Yield_k/h" (maize yield in kg per harvested ha)." If we examine the characteristics of other counties, we see that all counties on our map have the same fields, but they differ in relation to their values for each field. For example, while all the counties have the field "Prod_ton," each county can have a different production level.

By looking at the map, however, we do not know anything about the spatial distribution of values for fields such as "Prod_ton" and "Yield_k/h."

We can classify and map areas according to their respective values for a particular field, such as "Prod_ton."

The manner in which we display values in a map depends upon their characteristics.

Double-click on the theme name "Uganda_maize.shp" in the Table of Contents to bring up the Legend Editor dialog box again.

Let us look at the Legend Editor dialog box closely. There are two small boxes at the top entitled "Theme" and "Legend Type," each with a small button at the right with a downward-pointing arrow that is used to bring up a drop-down list.

Right now, the "Theme" in the Legend Editor dialog box is "Uganda_maize.shp" and the "Legend Type" is "Single Symbol". It is "Single Symbol" now because we have just one color (here light purple) representing the theme "Uganda_maize.shp."

Click on the down-arrow next to the box for "Legend Type" to bring up the drop-down list.

There are several choices for legend types in mapping in ArcView besides "Single Symbol."

  1. Graduated Color is used in maps to represent interval or ratio level data in different colors, with the color shadings displayed being related to the values represented.
  2. Graduated Symbols (not displayed in the drop-down list above) portray the sizes of points and lines in proportion to the values being represented; the colors of the symbols are uniform, only the sizes of the symbols vary.
  3. Unique Value maps assign a different color to each particular discrete category, name, type, or condition.
  4. Dot maps display differences in values in polygons by varying the densities of dots displayed, with the number of dots shown being proportional to the values in the classes being illustrated.
  5. Chart maps incorporate charts into the maps to display variations in values at particular places.

The same variable can often be represented in a variety of ways. For example, pest populations can be illustrated by all of the above types of maps.

Graduated Color Maps

To produce Graduated Color maps, select "Graduated Color" from the "Legend Type" drop-down list. We will produce a map showing the spatial distribution of total maize production for the various counties.

Note that when you selected "Graduated Color" from the drop down list, the Legend Editor dialog box changed with new choices appearing.

Two new boxes can be seen, "Classification Field" and "Normalize by" and both have the word "None" inside them. There is also an empty table below them, which we will use shortly.

The choices for "Classification Field" consist of all those fields that can be mapped in a graduated color map.

Under the category "Normalize by" you can divide each value by either the value of another field or by the total value of all items in the field (percent of total). For example, if your field has data on population, you could divide each value by another field that has information on total area to produce information on population density.

Move your cursor to the arrow pointing down on the right side of the "Classification Field" box and press on the arrow to reveal the drop-down list and scroll down and select "Prod_ton" from the list.

The empty table in the Legend Editor dialog box is now filled with five categories of maize production.

ArcView classified all the values in the field "Prod_ton" into five different classes (in tons of production) using a default of "natural breaks" in the data. For example, in this classification, the last (or fifth) class in the table, represented by the darkest red color, includes counties with maize production levels between 39001 tons and 71000 tons.

(Note: You can use other classification schemes as well. To do so, click on the Classify button in the upper right in the Legend Editor dialog box to bring up the Classification dialog box.

Here you can change the "Type" of classification by selecting a different choice from the drop-down list.

You can also change the "Number of classes" to be displayed by selecting the number of classes that you want in this window. The default for the number of classes in ArcView is 5.)

The five different production classes will be automatically assigned distinctive color symbols of varying shades of red because ArcView chose the "Red monochromatic" scheme to classify the data.

The name of this color scheme is listed at the bottom of the Legend Editor dialog box in the small box to the right of the category "Color Ramps."

Look at the table in the middle of the dialog box. There are two column headings, "Value" and "Label."

The "Value" column contains the values that ArcView uses to divide your data into classes. You can also change the values in the classes to produce your own categories. For example, if you were examining slopes, you would want "Value" categories such as 0-10, 11-20, 21-30, etc., not categories such as 0-8.654, 8.655-11.124, etc. that ArcView might produce.

The data in the "Label" column are what will appear as labels for each class in the Table of Content (its legend).

We should provide better labels for the data in the Table of Contents, such as adding commas and the word "tons." This would give a more professional appearance to our map, so go ahead and do this step by typing them in as I have illustrated below in the "Label" column.

(NOTE:When you click on the data in one of the labels to edit it as instructed above, the entire label is highlighted in black. Do not start typing right away as this will delete the data in the label. To edit the label without deleting the data, click on it a second time to remove the black highlighting!)

Your Legend Editor should look like this when you finish.

(Note: You could also change the Labels to non-number categories such as "low," "medium," etc.)

You can also change the color scheme used to represent the five classes by using the "Color Ramps" at the bottom left of the Legend Editor dialog box. Click on the down arrow to the right of the "Color Ramps" box (which now has "Red monochromatic" in it) to reveal the drop-down menu and scroll down and select "Purple monochromatic".

You can experiment with other color schemes available under "Color Ramps."

(Note: you can also change the color of any of the five symbols individually by double-clicking on its color under the heading "Symbol" and using techniques discussed in the previous Step.)

Now click on the Apply button on the bottom right of the Legend Editor dialog box to apply this new color classification scheme to our map. Close the Legend Editor dialog box to reveal the new color scheme on the map of Uganda, which should now look like this, with the spatial distribution of maize production levels in counties clearly illustrated.

Remember that before in the Table of Contents we just had the title "Uganda_maize.shp" with no other information. Now the five classes of production levels have been added below the term "Uganda_maize.shp," with both the five colors and the five categories of production levels in tons.

Note: on your screen, you sometimes cannot at first see all the numbers and labels in your five classes of a theme in the Table of Contents as you see here. If you want to reveal the complete numbers and labels in such cases, you need to drag the right hand border of the Table of Contents to the right with your cursor.

There is still one problem here. The current theme title--"Uganda_maize.shp"--no longer reflects the new categories that we have created. We need to change the title of our theme in the Table of Contents to "County Maize Production" to reflect the new values.

To change the title, go to the buttons at the top of the screen and click on the Theme Properties button

THEME PROPERTIES

which brings up the Theme Properties dialog box. First. click on the Definition icon on the left. Your dialog box should look like this:

At the top, where it says Theme Name it now has "Uganda_maize.shp." Delete that title with your cursor and type in "County Maize Production."

Then click on the OK button and notice that the new theme title appears in the Table of Contents.

Unique Value Maps

We are now going to create a map that displays unique values. We will first use data on the district names from the theme "Uganda_district.shp." Each district name is a unique value.

Display only the theme "Uganda_district.shp." Make sure that the box next to the theme name "County Maize Production" is unchecked in the Table of Contents so that the theme is no longer displayed in our map window.

Double-click on the theme name "Uganda_district.shp" in the Table of Contents to bring up the Legend Editor dialog box and click on the drop-down list for "Legend Type" and select "Unique Value" from the drop-down list.

Notice that a new category--"Values Field"-- has appeared in the dialog box.

From the "Values Field" drop-down list, select "Adm1" (first-level administrative unit).

The Legend Editor dialog box should now look like this.

Note that in the table in the dialog box there are several categories. We could change any Symbol individually, as discussed previously. We can edit items in the category "Label" just as we did above (in this case, though, it is not necessary). There is also a column labeled "Count" which provides information on the number in each category. We could also change the "Color Scheme" specified in the bottom of the Legend Editor dialog box if we wanted, as we did previously.

Click on the Apply button and close the Legend Editor. Change the title of our new theme in the Table of Contents to "District Names."

Our View should look like this:

(Note: When you create a legend based on unique values in the Table of Contents and the legend has many categories, it can become very cumbersome to leave all the categories displayed in the Table of Contents when you are no longer displaying that theme in your map. A long list makes navigating in the Table of Contents difficult. To make working in the Table of Contents easier, there are two things that you can do.

One is to open up the Legend Editor dialog box and change the "Legend Type" to "Single Symbol." Do that with the theme "District Names."

The other is to delete the theme from the Table of Contents. When you delete a theme, you do not remove it from your computer hard drive. You only remove it from the Table of Contents. To delete a theme, make the theme active and then go to the menu at the top of the screen and select Edit and then Delete Themes from the drop-down list.)

We can create another example of a unique value map for soil types.

Uncheck the box for the theme "District Names" and display the theme "Uga_soil.shp," which should be near the top of your Table of Contents. (If you had deleted "Uga_soil.shp," add the theme again.)

Double-click on the theme name "Uga_soil.shp" to bring up the Legend Editor dialog box. Click on the drop-down list for "Legend Type" and select "Unique Value" from the drop-down list.

In the drop-down list for "Values Field," select "Type."

Click on the Apply button and close the Legend Editor. Your map should look like this:

Each of the soil classes is displayed in the Table of Contents.

Zoom in to an area to see the spatial pattern of soils more closely.

Then identify the soils' characteristics. What useful fields are available?

Graduated Symbol Maps

Now we are going to employ another method of displaying interval and ratio level data, in this case in the form of point data. We are also going to combine such data with a polygon theme--in this case, our soil types.

Add the theme "Pests1.shp" to the Table of Contents. Assume that this theme has data on pest counts in traps at different locations. Then make this theme active and click on the Zoom to Active Theme(s) button.

You should now see the points representing the theme "Pests1.shp" on top of the polygons representing the soil types.

Identify the fields in the point features on our map.

Note the field "Pestcount1" and the value associated with it. Each point has a different pest count associated with it. The value at each point is different.

How do we display the spatial patterns in the values?

Close the Identify Results box.

We can use graduated symbols that display circles of varying size at each point in proportion to the value at each point.

In this case, the graduated circles will represent the number of pests at each point.

Double-click on the theme name "Pests1.shp" in the Table of Contents to bring up the Legend Editor dialog box.

For "Legend Type," select from the drop-down list "Graduated Symbol." For "Classification Field" select "Pestcount1."

Notice that the different categories of values are represented by circles of varying size but similar color.

(Note the category "Symbol" at the bottom left of the Legend Editor dialog box and the box to its right with the color of the point symbol (here blue). You can double-click on this point symbol to change its color, as we did in the previous Step.)

Click on the Apply button and your map should look like this.

In GIS, we are concerned with understanding how one spatial pattern (in this case, of the pest counts) is related to another spatial pattern (in this case, of soil types). Here we can use our powers of visualization to examine the influence of soil type on pest counts.

Which soil type is most closely associated with the highest pest counts?

This is a very simple example of using a polygon theme and a point theme together to examine possible spatial relationships.

We can also change the scale of the graduated circles to make the map easier to interpret.

Double-click on the theme name "Pests1.shp" to bring up the Legend Editor again.

Notice at the bottom right of the Legend Editor is the category "Size Range:"

This section enables you to change the range in the sizes of the graduated circles displayed on a map. Right now the default sizes range from "4" to "12." Change the size "12" to size "20" by clicking on the down-arrow to bring up the drop-down list.

Click on the Apply button and your map should look like this:

Notice that the contrast in sizes among the graduated circles is now more apparent.

Now display the theme "Iganga-river_50000.shp" on the map. Remember that you must drag the theme "Iganga-river_50000.shp" above the theme "Uga_soil.shp" to see it in our View.

What is the significance of the river for the spatial distribution of pest counts?

Remember from the previous Step that we can make the line representing the river appear thicker so that it is easier to see in our View. Go ahead and change the thickness of the line representing the river.

Note that you can also use graduated symbols with line themes, such as roads and rivers. The thickness of the lines representing them would be proportional to their importance (for example, all-weather roads would be represented by thicker lines than would be the less important dirt tracks).

Here is an example of displaying different types of roads with graduated symbols (in this case, lines of different thickness) based on the importance of the roads, combined with different colors for added clarity. This is a map of the Iganga area.

We have learned how to map the spatial distribution of values.

Now we want to learn how to print such maps.

[Return to beginning]

[GO TO STEP 8]