Sunday, December 3, 2017

Food Deserts in Marion County!



The map above shows a section of Marion County, Florida that I am analyzing for areas considered food deserts. Food deserts are classified as areas where individuals do not have immediate access to healthy foods. This map was created using open source software such as QGIS, ColorBrewer, MapBox and then uploaded onto the web using Leaflet.

I obtained the data for this analysis from the Marion County Open Data Portal. I also obtained population information from Factfinder.census.gov. The population table was then joined to the Marion County census tracts shapefile. The Grocery Stores layer was produced by creating my study area file in ArcGIS, converting it to a KML file and uploading it into Google Earth. From here I searched for grocery stores within the area and created a point layer. I then took my Grocery Stores KML file and converted it into a Shapefile. I found using Google Earth to create my Grocery Stores file to be helpful because it is easy to search for businesses and the street view allows you to ground truth.

Below is a link to my web map, give it a look!

Marion County Food Deserts

Monday, November 27, 2017

Creating a Web Page Using Open Source Software

This week I created a web map using only open sourced software. The map is of Pensacola, Florida and shows Food Deserts in the area. The creation of the map began by analyzing what census blocks within Pensacola were food deserts in QGIS and then transferring the newly created layers into MapBox. After choosing a style and creating a map in MapBox, Leaflet was used to publish the map on a webpage. 
Check out the link below for the finished product!

Friday, November 17, 2017

Using QGIS to Analyze Food Deserts




The focus of this lab was to become acquainted with QGIS, an open sourced mapping software available for free on the web. The objective was to look at grocery stores in Escambia County, Florida and analyze what census tracts are considered "food deserts" or areas without grocery stores close by. 

After becoming so familiar with ArcMap I was not looking forward to being introduced to a new software but after playing around with QGIS I have come to find that it is very intuitive and easy to use. I find it exciting that this software is free and provides many of the same functions that ArcMap possesses. 


Tuesday, November 14, 2017

Supervised Classification of Land Use/Land Cover

This weeks lab focused on using Erdas to perform a supervised classification of land use in Germantown, Maryland. A supervised classification automatically classifies an image by using training sites from multiple spectral bands associated with land cover type to guide the computer's classification. 

Tuesday, November 7, 2017

Unsupervised Classification using ERDAS

The map above was created by performing an unsupervised classification using ERDAS Imagine. The image was classified by assigning pixels to different classes and then transporting the final image to ArcMap where a final cartographic product was produced. 

Friday, November 3, 2017

Ordinary Least Squares (OLS) Regression of Meth Labs in West Virginia



This weeks lab focused on analyzing data prepared in previous weeks. This was the first introduction into statistical analysis of spatial factors. The purpose of this lab was to use the Ordinary Least Squares (OLS) regression tool in ArcMap to analyze 29 independent variables to see what factors may be a precursor to meth labs (dependent variable). By using census and education data, the tool was run and the results were analyzed. After analyzing the results, variables were removed or added one at a time to determine which explanatory variables were significant. After running the tool 29 times, starting completely over, and running it 31 more times, the table and map that are shown above were produced. 

Tuesday, October 31, 2017

Finding Heavy Vegetation by Manipulating Band Order


This image was manipulated to have a band combination of R6 G4 B7. This made the heavily vegetated areas show up bright green while more populated and developed areas appear pink. This allows for analysis of vegetated areas to become much easier than if the image were true color.