Brevard County Food Desert - Palm Bay & West Melbourne

This has been a crazy-fun experience. My last project for GIS 4930, Special Topics in Geographic Information Systems is to create a webmap using Open Source GIS tools. The subject of this last project continues to be Food Deserts. For my project, I selected two cities in Brevard County: Palm Bay and West Melbourne.I used QGIS (Quantum GIS) as a FREE and Open Source, ArcMap -substitute to conduct most of my analysis. I had to use ESRI's ArcMap for the Near tool to compute where the Food Desert and Food Oasis existed relative to my cities above. After running the tool in ArcMap, the QGIS table looked like the below:



I used the 2010 US Census data. The first red circle represents the Census Tract information; the second circle is POP2010. I used the tract information to compute the centroids of the population so that I could measure a Euclidian distance of one mile from the grocery store layer. The last column is the Near_Dist data derived from ArcMap – Like I said, I could not use QGIS to do everything I wanted – This Near column is the result of executing the Near Tool with a one mile search. Any item that returned a -1 is farther than one mile from the grocery store and hence in a Food Desert by my definition.

      

I used Google Earth to collect the Grocery Store information and the US Census data to compute population centroids. I used a number of FREE Open Source GIS tools, such as QGIS, Mapbox, and Leaflet, and I gained valuable experience doing so. Below is my QGIS and Mapbox maps.

My Mapbox Link: Mapbox Brevard Food Desert

The data I created is very good and highly useful for the purpose of determining Food Deserts. I believe that my data fairly represents the Food Desert situation in my study area.

Here is my final presentation for GIS 4930:

Link to My Presentation: My Brevard County Food Deserts Presentation

It has been a pleasure conducting this brief study and I look forward to many more years of using free and open source tools for GIS analysis.

Mapbox and Brevard County Food Deserts

This week we continue work on our Food Desert maps using open source GIS tools. I learned an incredible amount about Mapbox and Leaflet, but I am no where as comfortable with these Open Source tools as I am with ArcMap. Nonetheless, this has been a great learning experience.

I used data from Google Earth to collect and compile my grocery stores layer and I used the population centroids, based on the US Census data for both Palm Bay and West Melbourne. The US Census data indicated that the census blocks were farther apart as you traveled south from the center of Palm Bay. This meant to me that the population was much less dense. As I reviewed the aerials of south Palm Bay I discovered this area is very rural. If I did this lab exercise again, I would use a greater distance from a grocery store, perhaps five miles, for south Palm Bay, however, I would have to determine a definition to confidently say where Palm Bay turns rural.
I gained valuable experience using TileMill and then loading my Tiles (file extension = .mbtiles) to Mapbox.  However, the link above tells us that TileMill is no longer in active development; looks like I'll be using Mapbox Studio Classic instead.As an Open Source program, Mapbox is constantly changing, so this made the lab a little more challenging. In the end I was able to compile a Mapbox Dataset that had a basemap and both my Food Desert and Grocery Store layers. Below is what my map looks like at this point.

As you can see, there appears to be a no grocery stores south of Malabar Road. That is why so much of Palm Bay is a Food Desert. I suspect that more grocery stores may move into south Palm Bay in the future.

Open Source tools and Food Deserts

What a week. Happy after Thanksgiving to everyone. This week was about open source GIS tools. I learned about TileMill and Leaflet. Great tools for building web maps. I had a challenge to get all the tools on Leaflet to work properly for me. Tools such as the Layer Control or Geocoder. Leaflet is an html based program that allows you to upload maps using Tiles (Tilemill) and customize them by essentially copying the html code. However, even individuals that have extensive experience with code struggle with getting the correct, code, spacing, punctuation, and the case of letters. Needless to say, this week was a challenge. Given more time and coffee, I am sure I will enjoy using both TileMill and Leaflet in the future.

The objective this week was to develop a map showing Food Desert locations surrounding Pensacola Florida. Below is the link to my Food Desert map. It is a work still under construction. Please check back later to see what features and tools I have added.

Pensacola Food Desert

Open Source GIS: Food Deserts

This week we are working with an open source GIS tool called QGIS (Quantum GIS). Open source programs are a great way to work as a community toward a common goal. This week we are preparing data to compile a Food Desert map of Southern Escambia county. The final project will be to create a Food Desert map of an area of my choice. A food desert is defined as an urban area with a grocery store or fresh food source, that is greater than 1 mile away and in a rural area, a grocery store located greater than 10 miles away. I took the data from Escambia County and the US Census data for this area and applied a layer for grocery stores. Below are the Food Desert maps I produced. 

The first QGIS map shows Escambia County in relation to all Counties in Florida. 

  The second map created in QGIS shows the relative  location of Escambia County and the Study Area in the inset, and the two data frames with the study area that depict the Food Desert and Food Oasis locations:

Statistical Analysis:Ordinary Least Square – Meth Lab Indicators

This week we continue to work on our project of Statistical Analysis. I used ArcMap and the Spatial Statistics tools available to me: ArcToolbox, > Spatial Statistics Tools> Modeling Spatial Relationships> Ordinary Least Squares. Regression analysis is the most commonly used statistic in the social sciences, hence I used the OLS tool for this study. In this lab exercise I am examining the relationship between the existence of Meth Labs (Meth Lab Density), my dependent variable, and certain socio-economic factors. I ran the Ordinary Least Square tool in ArcMap more than 22 times to remove extraneous variables from the 2010 US Census data so that I could finally arrive at the six explanatory (independent variables) that I will use in my study report.

My explanatory variables are listed below:

The objective of running the OLS tool so many times was to create a valid model that would aid in the prediction of the presence of Meth Labs. This involved examining a variety of statistics, listed above, in a methodical way to evaluate six separate Checks:

     1. Are your independent variables helping or hurting?

     2. Are the relationships what I expected?

     3. Are there redundant explanatory variables?

     4. Is your model biased?

     5. Are any important independent variables missing?

     6. How well is the model predicting the dependent variable?

After reviewing each "Check" above and after sifting through approximately 70 US Census attributes, I arrived at my destination with data that still needs to be evaluated. But, I can say at this point that I have six explanatory variables that show some relationship to the prediction of Meth Lab Density.

Statistical Analysis with ArcGIS - Meth Labs in Charleston, WV

This week was "Prepare" week for Statistical Analysis with ArcGIS. Below is my basemap of the two counties in West Virginia we will be examining using Statistical tools in ArcGIS. The map depicts Putnam and Kanawha counties and a simple distribution of Meth Lab Seizures, courtesy of the DEA and the National Clandestine Laboratory Register.

I took 2010 US Census data for this area and computed percentages for various categories, including, single female head of household  (pcnt_FCHLD), My hypothesis is that where there is a high (or elevated) percentage of single female head of household, there will be a corresponding high probability of a meth lab. I will test this hypothesis using ArcGIS Statistical Analysis tools. There are various other conditions that make it likely for a meth lab to exist, however, pcnt_FCHLD is the primary attribute I will use in my study. It is my goal that the results may be useful to DEA and local authorities whose objective is to seize Meth Labs and arrest anyone involved in the production of Methamphetamine.      

Mountaintop Removal:Report Week & MTR Story Map Journal

This week in Special Projects, we concluded our module on Mountaintop Removal (MTR) mining sites. We took a deep look at the data by processing Raster images of the Appalachian Coal Mining Region. To do this, we were split into several groups so that we could analyze Landsat images that would reveal MTR sites in this region. I was a member of Group 3 and I analyzed Landsat image  LT50190342010243EDC00. Specifically, I processed this image using ERDAS Imagine to re-classify areas that were MTR or Non-MTR and I used the Multipart to Singlepart tool to create several more polygons for analysis. After classifying the MTR sites using ERDAS Imagine, I conducted an accuracy check by generating 30 random points and comparing those points to a 2010 aerial image of the same region. My accuracy check yielded 28 out of 30 points that were True or correct which gave me a 93% accuracy. My results were a slight decrease in the total acreage of MTR mining sites. Though this is possible due to restoration, the decrease in amount of MTR acres does not provide remediation for the source of drinking water for the families who live in this region. There is a correlation between MTR mining and poor health, fish kill, lower bird hatchings, and poor water quality.

There are six stages to MTR mining: Clearing, Blasting, Digging, Dumping Waste, Processing, and Reclamation. During the blasting phase it is not unusual to cause violent shaking to families homes located miles away from the MTR site, nor is it unusual to find coal dust coated on the their homes. In my opinion, it is time to end MTR mining. As a nation, we have many other forms of energy available to us and we should make use of these alternate energy sources. The Coal Mining region located in the Appalachian Mountains would benefit immensely if we converted from MTR mining to wind, solar or other methods for producing electricity. Changing from MTR mining to solar energy, would be another great way to create jobs and provide America a lasting and renewable energy source.

Please see my Map Journal - MTR Mining: Streams and Basins

Mountain Top Removal: Analyze Week

This week is a continuation of Mountain Top Removal. The main objective for this week was to Reclassify an MTR raster. To accomplish this, we used a Landsat image and manipulated it using ERDAS Imagine. I also learned about SkyTruth, a non-profit organization that provides  remote sensing and digital mapping technology using satellite images and other visual technologies, to provide information and data to environmental advocates, the media, and the public.  

My task was to examine a digital multispectral image that represents a portion of our study area. Additionally, this week we signed up for small groups. I am in Group 3 and we have two Landsat images to work with:

LT50190332010243EDC00.tar

LT50190342010243EDC00.tar

Using ArcMap I was able to use the Training Sample Manager to sample various polygons from the image and classify them as MTR or NonMTR. This could be a tedious task but I found it mildly entertaining.  I created the following histograms and scatter plots to check the accuracy of my sampling:

 

Looks pretty good to me. After this, I used ERDAS Imagine to create an AOI layer and then use the "Grow tool" to select pixels that were in an MTR area. Conducting this analysis took time for many of the tools to process the data. Once again, this was a fun exercise even though I had to wait patiently between each step of processing data to arrive at my final product that I then exported back to ArcMap to complete.  Below is my final re-classified MTR image 

Mountain Top Removal and Map Stories

This week I worked on a map concerning Mountain Top Removal for coal mining in the Appalachian mountain region. The EPA defines Mountain Top Removal as:

“Mountaintop removal/valley fill is a mining practice where the tops of mountains are removed, exposing the seams of coal. Mountaintop removal can involve removing 500 feet or more of the summit to get at buried seams of coal. The earth from the mountaintop is then dumped in the neighboring valleys.” (ilovemountains).

I used several tools to make this map. The bulk of the work was in creating the Hydrology Dataset from the Mosaic Group Raster Data. There were plenty of steps involved including:

- The Fill tool — This step modifies the raster to prevent water flow from pooling.
- The Flow Direction tool — This process assigns each pixel a value representing the direction of water flow across each cell.
- The Flow Accumulation tool — For each cell in the raster, this tool calculates the total number of other cells that flow into it.
- Calculate 1% of the pixels in the image —  A generally accepted condition is that the Flow Accumulation value must exceed 1% of the total number of pixels in the raster. If more than 1% of the total pixels in the raster flow into an area, it is probably a stream.
- The Con tool — This tool defines the threshold of flow accumulation values that qualify something as a stream
- Stream to Feature — This tool creates polyline vector features from the Con tool output raster

This was a lot of work but I am glad I used all the above tools for the exposure and practice. Below is my basemap.

Another skill I worked on this week was to create a Story Map and a Story Map Journal. The subject for both stories concerned the 6 Stages of Mountain Top Removal:

- Stage 1: Clearing 

- Stage 2: Blasting

- Stage 3: Digging 

- Stage 4: Dumping waste 

- Stage 5: Processing

- Stage 6: Reclamation 

You can find My Story Map at: Mountain Top Removal

I also created a Story Map Journal, but this is still a work in progress. If you would like to take a look at it now, go here: Mountain Top Removal: Streams and Basins

Network Analysis: Report Week for Tampa Evacuation

It has been a very long week. Lots of excitement and drawing of Storm Drains with Brevard County GIS Dept. I am learning a great deal and keeping busy. For this Special Topics assignment, I completed my work on the Tampa Bay Evacuation project by compiling a Pamphlet that could be used to inform relatives and hospital patients of the Tampa General Hospital of critical information in the event of an evacuation. Below is one page of the pamphlet I created.

Another Network Analysis I performed was to compute multiple evacuation routes from downtown Tampa to Middleton High School Shelter. This involved setting-up the Network Analysis to load or account for fifteen different locations in the downtown Tampa area. Here;s my results:

Another Network Analysis I ran was to generate multiple polygons to find the closest shelter, there were three to consider, for residents in the Tampa area. The analysis was interesting  because I cannot remember using this tool before though I do recall using the Multiple Ring Buffer. This was better. Here's my map for the Closest Shelter:

Network Analysis Week 2 - Tampa Evacuation Routes

This week it was time to analyze our data from the Prepare Week for the Tampa Evacuation exercise. The objective was to find the closest shelter for residents in specific areas in Tampa. Additional objectives included finding the best routes to supply the Shelters from the Tampa Armory and where and how to best evacuate patients from Tampa General Hospital located on Davis Island.

I learned a lot using the Network Analyst Toolbar and the Network Analyst Window. I created a new Network Dataset and set-up the network using Time (seconds) as the Scaled Cost to be evaluated relative to the route selection. I also used the Flood Zones layer (grid code) to group the Scaled Cost to be evaluated in the route selection. I created additional polygons to form the three sections around the Shelter facilities where residents would evacuate to in the event of flooding. I believe this was the most important message to convey, not that a hospital evacuation nor delivery of needed supplies from the Armory aren't important, it's just that I felt informing thousands of potentially panicking people (the triple "P" threat) where to go for safe shelter seems more important to me. Below is the map I created this week.

GIS 4930 Special Topics: Project1 Network Analyst

I know I had a Summer break but where did the time go? It feels like it's been a very long time since I have done Lab Work, while at the same time, where did the Summer Break go to?

This week, I start GIS 4930 Special Topics, with a Network Analyst project. The project, is broken into three one week phases - good thing too.  So, this week, I begin with reviewing my data and commencing the "Prepare Week." The following weeks I will actually conduct the "Analysis" and write the "Report."

The scenario is that I have been hired by the city of Tampa Bay to conduct an analysis of roads that could be used for evacuation in the event a hurricane causes flooding. I will use the following information:

1. Classified DEM polygon layer (with appropriate color scheme)
2. Flood Zone layer with appropriate color scheme
3. Hospitals, police stations, fire stations, shelters, national guard armory
4. Roads

This week's objective was to organize and prepare by assembling and preparing all my data. I used a Clipping tool (MassClip) and Re-projection tool (MassProject) and I re-classified a DEM and converted it from a Raster to Polygon. The Python work was very cool as I just finished GIS 5103, Python Programming; I could actually understand and appreciate the Tools written using Python. The last part of the exercise was to make a map package and share my map with a colleague. This included writing an email to let this person know what I had done and solicit feedback. I think writing the email was a challenge since I did not know the level of detail required for this fictitious colleague. Here's what my "Prepare map" looks like at this point. I can always spend much more time on any lab and I am sure I will tweek this map as I conduct the next Phase of Project1 - Analysis Week.