Meth Lab Analysis Week

Project 3 - OLS Analysis

GIS 4930

This week we delved into the world of Statistical Analysis in GIS.  Specifically the Ordinary least squares and geographic weighted regression.  Below is the OLS table and the map showing the results of that OLS analysis.  I followed the lab, and feel that my results fell within the directions given.  As this type of analysis is widely uses, I would like to have a more in depth understanding of it.  Statistical analysis is not a strong point of mine, and I felt I need to understand it better.

The map itself turned out quite well and produced the results I theorized it would.  Meth labs would be more likely to appear inside the urban populated area of Charleston WV.  Rather than show the Standard Residual and the decimal breakdown, I took the time to make it more readable to my viewers by describing where there may be trends in the legend.  I felt showing a digital values limited your audience, as most people would not understand what the Standard Residual values mean.  As this map targets law enforcement and not my GIS peers this only makes sense.  Also, I left the Inset to the Charleston area as this is the focus of most of the meth lab increase likeliness.  

Overall, this project was a bit difficult to understand and move forward on.  I would like to have more time to focus on understanding this type of statistical analysis in a more confined study with limited variables so I can see and understand the significance of the statistical changes.  Although advanced, I did understand the basic principles, but not to the level I feel I should. 

Project 3: Statistics - Prepare Week
Meth Labs per Square Mile - Charleston, WV

GIS 4930

For this weeks lab we began the Prepare Week by getting a basemap together that will be utilized in next weeks lab.  I identified the Charleston, West Virginia area as our Study Area for the project.  For this project we are looking into the socio-economic factors for the distribution of methamphetamine labs.  Utilizing the Census Tracts data from the U.S. Census Bureau and merging that data with education and lab locations my hope is to be able to identify areas that may show a greater likelihood of having an illicit meth laboratory.  

Some of the steps to produce my final basemap included calculating multiple percentages.  I utilized Field Calculator to calculate percent population growth, percent white, percent roommates.  Later in the lab we sped up the process of calculating percentages by using a python script.  This script quickly calculated multiple attribute fields and I was tasked with modifying the script to include percentages for age groups 40-49, 50-64, and above 64 years.  I also wrote script to calculate the male to female ratio and percentage of uneducated individuals in the population.

Utilizing a Spatial Join, I was then able to get a total count of meth labs per census tract.  This in turn allowed for me to calculate the number of labs per square mile for any specific census tract.  I used this data to then show a categorized symbology with a color ramp.  This symbology showed areas where there was meth labs and how abundant they were in that specific tract compared to others in the study area.  My final task was to clean up my attribute table, removing unneeded fields and produce a basemap for this weeks project.  

Project 2 - Report Week

Mountain Top Removal Analysis

GIS 4930

Mountain Top Removal (MTR) was the focus of this project as well as the environmental impact it has on the countryside.  Utilizing Satellite Imagery from 2010, I was able to do a multi-band spectral analysis on those areas in our group study area (Group 3) and determine locations that may have MTR projects ongoing.  I also utilized Digital Elevation Models (DEM) to run analysis and determine locations where streams may be located.  Streams locations are important as these are greatly impacted by MTR projects.  Runoff into these streams could have detrimental impacts on the environment and surrounding wildlife.

Starting off the project we identified what groups we would be in, and those groups were assigned specific study areas.  Utilizing DEM’s we not only created a steam layer, but also identified watershed locations.  Later in the lab, the group was tasked with identifying MTR locations using 2010 reclassified satellite imagery.  This was done in ERDAS Imagine program.  The classified locations were then brought into ArcGIS and the data was converted from a raster to a polygon feature.  With the polygon feature identified noise and interference was removed.  This consisted areas around Roadways, Highways, Streams and Major Rivers, all of which shared a similar bands with MTR locations.  Finally with this removed we isolated those polygon features that were larger than 40 acres.  This left us with the most likely polygons that were MTR locations.

Our group consisted of four members.  Each member played a role in the production of the data.  We had two satellite images to analyze so we assigned 2 users to each.  I took on the responsibility of keeping communication open.  I started many of the discussions where we needed to communicate back and forth to share information.  I also was able to return initial results to the group quickly so I could share my results and give help on issues that may have arisen.  The final MTR features were posted to ArcGIS Online to share the results that the group found.  I added another layer to the map that shows coal fields as identified by the USGS Eastern Energy Resources Science Center.

Mountain Top Removal Map

ArcGIS Online Login Required