GIS In Archaeology

Lab Exercise 8

Step 1: Create a new map document
Start a new map in ArcMap.
Expand E:\GIS_In_Archaeology\Data\USA\Pennsylvania folder and add the layers Pop_Places and Counties to the map document

Step 2: Classify Populated Places
Currently there is no distinction between the different populated places. They are all displayed with the same symbol.

Right click on the Pop_Places layer in the Table of Contents and bring up the properties window and select the symbology tab. Select the Graduated symbols option under Quantities and select Pop_98 as the value field

Notice that currently the classification scheme for the layer is Natural Breaks (Jenks) using 5 classes.
Click the Classify button to bring up the Classification window. Modify the number of classes to be 6.
Notice that currently the Break values between classes are currently set up as 5726, 20512, 52951, 102640, 340520 and 1436287

Step 3: Modify Break Values
A printed out map might be somewhat easier to interpret if the groups had more rounded values. Change the break values by clicking on the entries under break values and typing in new numbers. Type in the values 9999, 24999, 49999, 99999, 499999 to modify the classification scheme. Notice the classification method is now listed as Manual since we typed in break values.


Step 4: Modify Point Symbols
Apply the new symbology classes to the map. The map now distinguishes between towns by using larger symbols for places with larger populations. Nevertheless the use of a monotone coloring sheme still hampers the interpretability of the map.

Bring up the symbology tab for the layer. Modify the overall symbology by changing the replacing the current symbol size range (currently it is 4 to 18) to be 3 to 16. Click the Template button to bring up the symbol selector window and change the template symbol to be Circle 2.


Modify the symbol for the topmost population range (501-9999) by double clicking on the current symbol to bring up the Symbol Selector window. Modify the symbol to use a gray color and size 3.

Modify the rest of the entries to match the specification below:
Symbol Range Color Size
501-9999 Gray 3
10000-24999 Blue 6
25000-49999 Green 8
50000-99999 Yellow 10
100000-499999 Orange 13
500000-1436287 Red 16

The combination of both color and size variation helps make the population distribution more easily interpretable.

Step 5: Map the County Population Distribution

Turn the Pop_Places layer off so that we can concentrate on looking at the county map.
Right click on the Counties layer in the Table of Contents and bring up the properties window and select the symbology tab. Select the Graduated colors option under Quantities and select Pop2001 as the value field

Looking at the map the counties with the highest 2001 populations are generally in southeastern Pennsylvania as well as in western and northwestern Pennsylvania.

Step 6: Modify Break Values
Return to the symbology tab for the counties layer. Notice that currently the classification scheme for the layer is Natural Breaks (Jenks) using 5 classes. Click the Classify button to bring up the Classification window. Notice that currently the Break values between classes are currently set up as 94174, 252206, 475156, 758265, and 1509243.

A printed out map might be somewhat easier on the eyes if the break values created more rounded values (99999, 249999, 499999, and 749999). To change these values click on the entry under the break values window and type in the appropriate numbers. Notice the classification method is now listed as Manual since we typed in break values manually.

Apply the new symbology to the map to view the effect of the changes.

Step 7: Data Normalization
Looking at the map of Pennsylvania there are obviously differences in county sizes. The current map shows just the distribution of population but perhaps we’d be interested in viewing the distribution of population density.
Bring up the county layer symbology window once again. Notice that in addition to a Value field there is also an entry for Normalization. When a normalization field is specified, ArcGIS will divide the values specified in the Value field by the values specified in the Normalization field. To mitigate the possible effect of county size on population (perhaps some counties have larger populations simply because they are larger) specify the field Area as the Normalization field.

On the resulting map we can see that it does appear that county size may have exerted some influence on the distribution we saw in the original map of county populations.


Another question of interest may be attempting to view population trends. Instead of looking at a map that just shows the distribution of population in Pennsylvania perhaps we’d like to see which areas are experiencing population increase and which areas are not.
Return to the symbology window for the county layer. We will once again use the Pop2001 field as the value field but change the Normalization field from Area to Pop2000. Values over 1 indicate counties where population grew from 2000 to 2001, while those less than 1 indicate counties of population decline.

The map resulting from this procedure is markedly different from the previous maps. What are some trends evident in this new map?

Step 8: Classification Conflicts
Turn the Pop_Places layer on once again. While the color scheme we selected for the Pop_Places layer in step 4 helped us to see distinction in city sizes it is now somewhat more difficult because the county layer is using a classification scheme using the same color pallet as the one we specified for Pop_Places.

To improve the resulting map it is necessary to modify the classification scheme for either the county layer, the pop_places layer, or both. In this case the big problem is that the colored point layer overlays a colored county layer with both using a similar color scheme. Given the range of colors in the point layer I decided to use a grayscale pallet for the county layer.
Bring up the symbology window for the county layer and modify the Color Ramp to be one which goes from white to black.


My major complain with the resulting map is that the darks seem to overly dominate the map.

We can create a custom color ramp correct this situation. Once again bring up symbology tab. Double click on the black box next to the bottom entry in the legend and from the symbol selector pick the Color that is one shade less than the most black.
When you return to the symbology tab right click in the symbology area and select Ramp Colors from the pop-up menu.


This new color scheme seems to correct the conflict issues which reduced the visibility of the larger towns. However, the smallest towns current use a gray symbol which can fade in to the background in those counties where a similar shade of gray occurs. Modify the symbology of the smallest town to use the Color Blue with a Size of 2.5.