Remote Sensing

Agricultural losses from 2019 Missouri River flooding

This was the result of a project I did for my remote sensing class. In March 2019, heavy rains on top of a deep snowpack and frozen ground caused an extreme flooding event along the Missouri River. South of Omaha, NE the flooding continued throughout the summer, resulting in much of the floodplain being unusable for agriculture.
My goal for the project was to compare Landsat imagery from August 2018, March 2019 at the height of the floodings, and August 2019 to see how much of the land had changed from agricultural growth to unused or flooded land.

I used Erdas Imagine software to conduct a supervised classification of the three images. From the classification, I was able to compare the agricultural land lost to flooding.

Close Window

Computer Cartography

National Parks within 50 miles of Amtrak stations

Beatiful maps have always interested me, and as I have become more interested in the graphic design, I decided to audit a computer cartography class in the summer.
This map was a personal project I did one afternoon after taking the train from Chicago to Indiana Dunes National Park. On the train I began to wonder what other national parks would be accessible from an Amtrak station.

The national park and Amtrak data I first combined in ArcMap and completed a buffer of parks within 50 miles of stations and added a base layer of a terrain relief. I exported the results to Adobe Illustrator.
In Illustrator, I changed the fonts and placed the labels, adjusted the colors, and added a fill and inner shadow to the states.

Close Window

GIS III

Map of Historical Earthquakes in Italy using QGIS

This is a map that I made using QGIS for a class assignment. The data I used for this was a subset of historical earthquake data from around the world, and I decided to use the graphic design capabilities of QGIS to create a map with a historical style.

I used a European country basemap and combined it with a world DEM layer to give the map some topography shading. Additionally, I overlayed a vintage paper jpg image to give the map an antique look.

The graduated symbols were created with the feature blending option in QGIS and were given a drop shadow and inner glow to make them stand out.

I added some historical images like the ship, compass, and earthquake drawings to give is a more authentic feel. Finally, I created the title banner and legend using Photoshop to add some decorative text.

Close Window

GIS for Urban Studies

Urban Agriculture Distribution in Chicago

Urban Agriculture is another one of my interests, and I hope it can improve cities by educating people about sustainable eating practices.

For this project I used Census data of low-income households which I mapped by Chicago community boundary and converted to raster format.
I then created a raster density map of a point shapefile of grocery stores inside city boundaries. I overlayed these two maps to create a "risk" map which showed what areas of the city might be at risk of being considered food deserts.

Finally, I added a map of community boundaries in Chicago and added the number of urban farms and community gardens in each community based off of the Chicago Urban Agriculture Mapping Project website and converted this to raster.

For the analysis, I combined the "risk" map with the urban ag map and found which needy community areas were well-served by current urban agriculture projects and which communities could benefit most from future projects. The map here shows the final results by census tract.

Close Window

GIS II

Ski Areas near Amtrak stations

This project came about as an intersection of two of my interests: Train travel and skiing. My goal was to see what ski areas would be located near Amtrak stations and accessible by train.

To complete this project I first found ski areas within 50 miles of Amtrak stations and within 1000 miles of Chicago (a requirement only for me as I didn't want to spend more than 20 hours on the train).

My main criteria after distance from the train stations was average yearly snowfall. I wanted locations that received more than 60" of snow per year. This data I found came in a CSV file with data from National Weather Service recording stations, so I first had to map the points using X-Y coordinates. Then I used Interpolation to fill in the gaps between recording stations and create the snowfall raster file.
Finally, I added a raster file of land cover data.

With the snowfall raster and the forested area raster, I was able to combine the scores into a final suitability map based on how forested each raster cell was compared with how much snow it received.

The final ski areas were only located in Colorado and the Northeast, with the Colorado ski areas coming out better ranked due to their higher snowfall totatls. The final Colorado map is seen below.

Close Window

GIS I

Midwest Campgrounds for Viewing Aurora Borealis

This was my first major project using GIS. My goal was to complete a Multi-criteria Evaluation of Midwest campgrounds that have the best characteristics for viewing the Northern Lights. I chose this topic as this is something I would like to photograph.

The following criteria were used to determine the best campgrounds for me:

• Within 300 miles of Chicago
• Within the KP level 7 or less
• Campground with a light pollution number
  of 3 or less on the Bortle scale
• Preferences: Open all year, near a body of water

For this project I used point shapefiles of upper Midwest campgrounds, However, I could not find already-made shapefiles for Wisconsin and Iowa, so for these states I had to manually geocode the geographic coordinates of each of the state parks with campgrounds in order to be able to do my analysis.

The KP-Index step (seen in the map above) shows the areas of the Midweset which can see the Northern Lights during solar storms of various strenths. The higher the number, the farther south you can see the lights.
I hand-drew the KP-Index polygons by tracing over a JPG file I found of the index from NOAA and then selecting only campgrounds that were within the areas with higher probability of seeing the Aurora.

As I wanted only campgrounds which were in dark areas, I then compared remaining campgrounds with at dark sky map and created a coding system in the attribute table to note the light pollution level of each campground. Those with a level of 3 or less on the Bortle scale were selected (see map below).

Finally, I coded campgrounds based on proximity to water and updated the attribute table. For the final analysis I selected campgrounds by attribute and ranked the top 10 based on the original criteria.

Close Window