Thursday, December 14, 2017

Project 4: Open Sourced - Report Week

Last assignment of the semester. This is a continuation of Project 4 where we were tasked with presenting our data in a PowerPoint presentation. Here is a link to my published web map: PWestFoodDesert and my PowerPoint presentation: PuebloWest.

Monday, December 4, 2017

Project 4: Analyze Week 2

This week, we created a web map of our own choosing. I choose the town that I live in Pueblo West, CO. Here are answers to questions asked about the process which should give some explanation.

1.       Explain your data including the source, processing required (did you create it? How?), and information about the quality/credibility of your food deserts and grocery stores. 

I used the US Census website to gather most of my shapefiles. Because there are only two grocery stores in my town, I used the draw tool to create the grocery store shapefile. The food desert shapefile was made by joining the city boundary, census blocks, and population data. I choose to use census blocks, because the first map I created was with block groups and the entire town was a food desert. Knowing that not everyone in my town lives in a food desert, I decreased the size of the study units. I think this improves the credibility of the data. The grocery stores are credible because I have first-hand knowledge (in-situ) of the area.

2.       Describe the data being represented. Are there trends you can identify?  If you used your local town to map, did you expect to see food deserts occurring where they are?  Does the data surprise you?

This is my hometown so the data did not come as a surprise. Pueblo West is located about 15
       miles west of Pueblo Co, and is still kind of the Wild West. City planning is poorly conducted if at all.
       I think there was a land grab many years ago and Pueblo West has a more area then Pueblo. This has
       led to neighborhoods that are massive and most parcels are larger than an acre. What I am getting      

       at,  the town is spread out causing most of the population to live in a food desert. 

Tuesday, November 28, 2017

Mod 4: Analyze Week 1

This week we were introduced to Mapbox and Leaflet. I used Mapbox to classify data from the QGIS data that I created last week. Mapbox is odd in that I needed to use the QGIS data and break it down into individual layers to create a classified product. ArcMap does this all in one step. The outcome is the same and this completed the task in open source which is cool.
Next, I took my map from Mapbox and used it in Leaflet. Leaflet is an open sourced application that can publish interactive maps to the web. My map is a continuation of week 1 but has been published to the web with some added features. There is now a geocoding bar to locate points on the map. I also added a legend, pop up windows and areas to identify areas on the map (food deserts vs. non-food deserts. Here is a link to my map, enjoy.

Friday, November 17, 2017

Mod 4: Open Source, Prepare Week

This week we started the open sourced mod. To introduce us into the subject, we used QGIS, an open source GIS, to examine food deserts in southern Escambia County. Food deserts are areas that are greater than 1 mile from a grocery store, a place with fresh food, convenience stores do not count.

I have never used QGIS before. So far, I can say it has some cool features and limited functions in other areas. I really enjoyed the print composer, it seemed to make map making more convenient.

I am including a map of the food deserts in Escambia County, Florida. This map divides the area out by census blocks. Centroids were used to represent the blocks and the Near tool was used in ArcGis to determine the distances between grocery stores and centroids. Once I had this data, I could use the select by attribute tool to separate the blocks outside of one mile and the ones inside of one mile. Here is the final map, the legend tells which blocks are food deserts and which are not.

Tuesday, November 14, 2017

Mod 10: Supervised Classification

This weeks Lab was on supervised classification. Our assignment was to take a satellite image of Germantown, Maryland and classify it by its spectral signatures. My first task was to create spectral signatures and AOI features. Next, classify the image. Lastly, recognize and eliminate spectral confusion between spectral signatures. Here is my final map of Germantown after the classification.

Sunday, November 5, 2017

Mod 9: Unsupervised Classification

This week I performed an unsupervised classification of the UWF campus. I performed the classification in both ERDAS Imagine and ArcMap. Once I reclassified the image, I recoded it to simplify the image into 5 classifications, making it easier to interpret. Here is the final image with areas calculated both by hand and using the area calculator in ERDAS.

Friday, November 3, 2017

Mod 3: Stats, Analyze Week

This week was analyze week for our stats mod. We are trying to determine if there is a correlation between meth production and demographics. One way to do this is by using the ordinary least squares (OLS)  tool in the ArcMap. OLS can determine if there is a connection in variables in the dependent and independent variables. Here is an OLS table that I created from 29 independent variables that I reduced to 6 and the associated map.