Friday, April 27, 2018

Module 4: Analyze 2 Fairhope Food Desert

I chose to focus my analyze to the area surrounding Fairhope, AL.  Any census tract that contained the Fairhope city boundary, I included in the study area.  Since some of the tracts extended much further than the Fairhope boundary, to fairly assess each tract, I located grocery stores and farmers markets for the entire tracts (not just within the Fairhope city limits).  I found the centroid for each tract and then did a Near Analysis to generate the near distance values for designating a food desert or oasis.  Once I joined this information to my tract layer, I then created a shareable and interactive map with Mapbox and can be found here.

This exercise also exposed me to tiling.  It involves rendering raster or vector data into smaller tiles that are easily displayed (ie load quickly) on webpages.

Thursday, April 19, 2018

Module 4: Analyze Week 1 Food Deserts

Since this week involved mainly coding, I found it very difficult.  I was unable to actually get the map to work.   We were introduced to two new programs:  Mapbox and Leafletjs.  Bot seem very useful and cool but I definitely left this lab slightly confused.  Certain components (such as the legend) on the web page are visible but the actual Pensacola area map from Mapbox is not working.

Here is the link to my non functioning web page.

Friday, April 13, 2018

Module 4: Prepare Food Deserts

The map generated this week displays food deserts and food oases in Escambia County, Florida.  A food desert is defined here as an area whose population must travel further than 1 mile to access a grocery store that sells fresh fruits and vegetables.  Census tracts were the defined boundaries and the centroid for the tracts was used as the point of measurement for the 1 mile radius.

This was my first experience using open source data and QGIS.  Regardless of its similarities to ArcMap, I still required the step by step instructions to navigate its tools.


Friday, March 16, 2018

Module 3: Methamphetamine Analyze Week

We performed an Ordinary Least Square Regression  (OLS) Analysis to predict the accuracy of meth lab locations in the Charleston, WV, area.  Through a 6 Check process, the social and economic factors discussed last week were eliminated or kept to enhance the accuracy of the regression model.  The final OLS chart is provided below and only uses 10 variables.


The model produced a density map that is represented  by the following Standard Residual map.  Standard residuals visualize the accuracy of the model's predictions.


Friday, March 2, 2018

Module 3: Methamphetamine Prep Week

This week began our analysis to correlate socio-economic factors with the location of methamphetamine (meth) labs in the Charleston, West Virginia area.  Factors such as employment, race, urban/rural communities, etc., have all been shown to be related to a prevalence in meth production facilities.  These factors can all be derived from U.S. Census data.
We were provided with a shapefile of known lab locations that have been "busted" and a shapefile of census tracts for the region with either the social factors given or that could be easily calculated.  We set up those census tracts with the required attributes and have outlined the study area as can be seen in the following basemap.  The final product for Module 3 will hopefully assist local law enforcement in targeting areas that have social and cultural predispositions for meth production.


Friday, February 23, 2018

Module 2: MTR Report



The final product from our last three weeks of work is a story map that walks you through Mountain Top Removal practices in the central Appalachians.  It also provides an analysis of a specific study area within this region with a breakdown of area effected.  The analysis can also be seen as a visual from ArcGIS online from this link.  My story map journal can be found here.





Friday, February 9, 2018

Module 2: MTR Analysis

This week's assignment created a raster that only contained MTR areas.  This was produced by compiling and running a supervised classification on Group 6's landsat imagery LT50180352010172EDC00.  The ERDAS Imagine supervised classification produced 50 classes that were later designated either as MTR or nonMTR areas.  Once this raster was opened in ArcMAP, a spatial analyst reclassification was used to create the image with only MTR areas included.

Below is a screenshot of the final product.