Course Materials for Geography 481, Project Three

Geography 481:  Intro to GIS
Project Three: Spatial Operations and Analysis

One of the most powerful features of a GIS (and one of the characteristics that distinguishes GIS from database management software) is the ability to select and evaluate phenomena based on their spatial (Geographical) characteristics.  

This project introduces techniques of spatial querying and spatial analysis using ArcMap software. In earlier exercises, selections and filters were based on values in the attribute database. Now, you will use spatial operations such as proximity buffers to produce subsets of your data. In the first part of the activity, you will select and analyze well water quality based on distance from a geographic point. In the second part, you will select and analyze air quality based on distance from a linear feature. Finally, you will produce a comparison of air quality based on distance from two different features.

Setup

Selecting Features Based on Proximity to a Point Location

In an earlier exercise, you used the zoom function to map TDS values for all wells located approximately 1 mile from the intersection of Azusa and the San Bernardino Freeway. Now, you will create a selected subset containing those same wells and compare their TDS values to those of the entire study area. As is the case with many of the operations in GIS, there are several ways to do this.

To select the wells you must first create a graphic circle with a one-mile radius centered on the intersection of Azusa and I-10.

Wells lying within the one mile radius should turn light blue indicating that they are selected. Now when you query the Wells layer, the results will describe only the selected set of wells.

Comparing Attributes of a Selected Subset to Attributes of the Whole

One reason for creating a features subset is to allow you to perform an analysis which compares the attributes of the smaller set with those of the larger group from which the subset was drawn (or to other subsets). Here you will use the Wells Attribute Table to compare TDS values for the two groups. Your goal is to measure whether the average water quality in the area of the selected wells differs from the average of all wells in your data set.

The count should indicate the small number of currently selected wells. Note the Mean TDS value for the selected wells.

Now you have statistics for the entire set of wells. Is the mean TDS for wells near Azusa and I-10 substantially different from the mean TDS for all wells? Is it higher or lower?

Close the table and delete the circle. (If the circle is no longer selected, you must use the "Select elements" arrow tool to select the circle before deleting).

Selection Based on a Buffer Distance around a Feature

In the previous example, you drew a circle around an X,Y location and used the resulting shape to perform a selection. Another common technique is to draw a proximity buffer around an existing geographic feature in one layer and use it to select features in a second layer. In the following example, you will select all air quality monitoring stations located within 3/4 of a mile of any freeway. You will then use the spatially selected set as a filter so you can produce a map of Nitrogen Dioxide (NO2) showing only those air quality monitoring stations lying within the 3/4-mile buffer.

For this portion of the exercise, you will be using the Air Quality layer instead of the Wells layer.  Through a fortunate coincidence, there happen to be air quality monitoring stations at each of the well locations so the two layers should look similar. To avoid confusion, you should make the Wells layer invisible by clearing the check box for Wells and make the Air Quality layer visible.

Keeping Track of Buffer Selections

As you know, the selected status of features is only temporary. If you want to keep track of which features satisfy certain proximity queries (Select by Location in this case), a simple modification to the attribute database will usually suffice. In this section you will select features that satisfy a specified criterion and then record this information as an attribute of those features.

The first step is to restructure the database to add a field that you can later use to store information on the buffer status (inside or outside) for each air quality monitoring station. First you'll create an integer field initialized to zero; then you'll replace the zero with a one (1) for those monitoring stations inside the buffer.

Note that the field is initialized with zeros, then close the table.

The next step is to select monitoring stations located within a 3/4 of a mile freeway distance buffer:

Before continuing, read through the selection statement. Think about what it is saying. Check to see that it will accomplish your goal of selecting all air quality monitoring stations located within 3/4 of a mile of any freeway.

Notice the selected points in blue. They should all be close to a freeway.

Next, open the table for air quality and calculate a value of one (1) into the Fwybuf field for each of the selected stations. There is nothing special about the values 1 and zero here. They are just a standard (and convenient) way to distinguish inside the buffer (1) from outside the buffer (0). They are nominal or categorical values.

Note: all stations inside the buffer now have Fwybuf = 1; all stations outside the buffer have Fwybuf = 0.

Now you can clear all selections. You won't lose the information identifying all air quality monitoring stations located within 3/4 of a mile of a freeway because you have added it permanently to the air quality monitoring station database.  In other words, you have used the spatial characteristics of various layers (in this case, proximity) to modify the attributes of the air quality monitoring stations.  By "hard-coding" this particular spatial relationship into the attribute table of the monitoring stations, you are now able to incorporate this relationship into your analysis.

Since the goal of this portion of the exercise is to create a map of NO2 showing only the air quality monitoring stations located inside the buffer, you must now use the value Fwybuf = 1 as a layer filter. When the filter is applied, only stations with Fwybuf = 1 will be visible; stations with Fwybuf = 0 will be excluded from further consideration and will be hidden from view.

The Air Quality layer will redraw showing only those stations which satisfy the condition Fwybuf = 1

Finally, create a map classifying Air Quality monitoring stations within the buffer into two NO2 categories: Below 150 and Greater Than/Equal to 150

As you admire your completed map, think about the sequence of operations needed to accomplish the goals of your analysis. You have successfully used spatial information to identify a subset of your data. In the first analysis, you compared the attributes of the distance-based subset to the attributes of the entire data set; in the second, you produced a map which filtered out all features not part of the distance-based subset.

Intellectual Interlude...

Spatial Join

Another useful spatial analysis technique is the spatial join, also referred to as join by location. Using this technique, GIS analysts are able to append attributes from one layer onto the attribute table of a second layer based upon some location criteria. This concept is also the basis for finding the nearest feature to another feature, counting the number of points that fall within a polygon, and determining whether features are inside or outside of a polygon.

Perhaps the most common use of a spatial join is to identify features that fall within a polygon. The simplest to perform and understand is the point-in-polygon analysis. Let's see how this works by figuring out which cities our various air quality monitoring stations are in.

Before you begin, remove the logical filter (Definition Query) so that all of the air quality monitoring stations are included in your spatial join:

All of the air quality monitoring stations should be visible.

Now:

The new shapefile should have been added to the TOC. Open the attribute table for the new layer you created. Scroll around and examine the new attribute fields that have been "spatially joined" to the city polygon layer. Compare the attributes to those of the original LA_County_City_Boundaries layer.

How many monitoring stations are in the City of Baldwin Park? What is their average NO2 value? How do these values compare to other cities?

... Back to our regular programming

On Your Own

Congratulations!  You have just landed a job at a consulting firm that monitors air quality in Southern California .  Your first assignment is to create a map illustrating the NO2 levels in the air near major freeways.  The 10 freeway and 210 freeway have different amounts of truck traffic and it is believed that the NO2 levels in the air will be affected by this difference.  Your new boss asks you to create a map and analysis to test this hypothesis.  

Answer the following: how does the average NO2 for air quality monitoring stations located within one-half mile of the 210 freeway differ from the average NO2 for stations located within one-half mile of the 10 freeway?

The new boss is persnickety so your map must meet the following minimum requirements:

FYI:  Note: to avoid confusion, you should remove any logical filters and return the Air Quality layer to Single Symbol before beginning to work on your map! You probably won't need the LA municipalities, either.

Software Note: For most operations, if features are selected then ArcMap will perform the operation on just the selected features automatically. However, for Select by Location you need to check the Use Selected Features box located under the Source Layer.

Snatched from the headlines: Not too long ago, most of the local news outlets were reporting the results of a recently published study that linked proximity to freeways and atherosclerosis. The article is linked here. Most of the coverage was on Valentine's Day. I am sure it was no coincidence that news about heart disease broke on the 14th.

Last modified 09/14/21