Lab exercise 3, due right before the class on Feb 11, 2008
EES6513: Advanced GIS, UTSA

 

Student name: ______________

 

 

Use Spatial Statistics

 

Purpose

 

            This lab is to help you understand the spatial statistics toolsets and their applications.

 

Step 1. Preparation

 

            Copy Lab3 under \\129.115.25.240\XIE_misc\EES6513\ to your local computer under the c:\EES6513\yourname or your portable driver. Open the .xmd file you used in Lab 2 and save it to Lab3. You will need some of the vector data you used in Lab 2 for this Lab. Since it is better to use the projection coordinates than the geographic coordinates in the Spatial Statistics tools, I suggest to reproject all your data into UTM zone 14N and WGS 84 datum.

 

Measuring spatial distribution:

 

Step 2. Add the plc00.shp shapefile from the Lab3 directory (reproject it before use it). This is the Census Designated Places Shapes as Defined in 2000 from the Texas State Data Center and Office of the State Demographer website (http://txsdc.utsa.edu/txdata/shapefiles/state_shape108.php). This Center is now at UTSA. This shapefile includes all the units (cities and towns) of Texas that have been surveyed and monitored for population changes.

 

Step 3. Create a new polygon shapefile (your study area) to cover the entire area of San Antonio to Austin corridor (see figure 1 below).

 

          

 

 

 

Step 4.  Use the Mean Center tool under the Spatial Statistics Tools to create a geographic center (saying geoc.shp) of your study area.

 

Step 5. Clip the plc00.shp to your study area and create a geographic center of this clipped place file (plc00_clip).

 

Question 1. How many records of the plc00_clip file has? does these two centers overlaying? why?

 

Step 6. We can also create a geographic center of each place (polygon) of the plc00_clip, using the same Mean Center tool, but select a Case Field of either FID or GIST_ID.

 

Question 2. what happen if you select other case fields such as PLACE and NAME?

 

Step 7. Now I want you guys to take this work one step further, similar as what we discussed in the lecture, to create a population center using the population of each place as the Weight Field.  The population data can be downloaded from the Table 2 at Texas State Data Center webpage: http://txsdc.utsa.edu/tpepp/2006_txpopest_place.php. In this table, you have 8 fields. what we really need are Place, 2000 Census Count, January 1, 2007 Population Estimate, and Numerical Change 2000-07. As a advanced GIS student, I hope you know how to get this data into a table format that GIS can use. If you really do not know how to do it, you can come to ask me. I will help you.

 

Question 3. What you want to do is to find the 2000 center and 2007 center and discuss if they changed or not, why and how?

 

(I hope you know to you need to link the population Table to the plc00 (or plc00_clip) shapefile, and you should use Join tool (one to one Join is the way to go). I have practiced this lab and I found a better way to do so. here are some hints: (1) You should join using either FID or GIST_ID, since they are unique to each place or polygon, while the place name will be used to help you to assign the FID or GIST_ID to the population table. (2) sorting two tables from a to z, using the plc00_clip as a starting point to find the same place name from the population table, so you can find and assign the FID or the GIST_ID to the record of the population table. (3) if you can not find a match from the population table to the plc00 table, you might just delete those records from the plc00 table. for a real project, you can not do this, you need to find where it is the problem. but for this lab, you do not have time to find the why there are errors and where they come from.)

 

Question 4. To find the most centrally located place (or polygon) from the plc00_clip, using the Central Feature tool. if you use population of each place (polygon) as the Weight Field, where it is the most centrally located place? and give a explanation or discussion.

 

 

Analyzing spatial pattern

 

Step 8. To see if there is any pattern of the population distribution for both the 2000 and 2007 in the study area, you should use the Spatial Autocorrelation tool as we discussed in the lecture.

 

Question 5. Show your figure of the spatial autocorrelation results (see an example below). Analyze and explain your results.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Question 6 (extra points of 10), if you do find cluster in the study area, please use the Hot Spots tool to identify where the cluster(s) is(are) located and give your analysis.