Advanced GIS

Advanced GIS Portfolio

arcgis raster tutorial

For this aspect of ArcGIS training, I worked with raster images in the program. Working with rasters is extremely important in remote sensing, which is the aspect of GIS that I enjoy the most!

arcgis spatial extension

For this training, I worked with the ArcGIS spatial extension, which enables all kinds of spatial analyses for rasters. ModelBuilder was also utilized here. The program was used to find the optimum location for a new school based on elevation and land use data, then find the cheapest way to pave a new road from an intersection to the school.

finding areas at risk of flooding in a cloudburst

I am also training in ArcGIS Pro-- This program was used here to analyze high tide in Venice.

This aspect of training utilizes digitial elevation models and complex models to perform flood analysis and determine which areas are more vulnerable to a flash flood.

using land cover to classify shrinking lakes

ArcGIS Pro was used here to analyze LandSat images of a lake in China to approximate how much it has shrunk. Analyses like these are my favorite things to do with GIS!

no dumping- drains to ocean

um-Dearborn campus map

It's time to put all this training to good use! Myself and three others were tasked with building an interactive map of parking spots at UM-Dearborn. We decided quickly that it would look better as an interactive map of the entire campus, including buildings. We used Google My Maps to create the map using layers digitized in ArcMap. This enabled the addition of pictures and better descriptive information.

I am pleased with how this map turned out. It is definitely much better than any previous map of campus!

analyzing earth's changing albedo

Finally, I designed my own research project using GIS. I decided to study the change in Earth's albedo, or ratio of solar energy reflected to total solar energy received. I used data from the EROS data center (which I happened to visit during Geology of the National Parks) sourced from NASA's Terra and Aqua satellites. Albedo is important when studying both the Earth's energy budget and climate cycles.

The results of this project come in three formats: The first is a presentation given on December 6th, 2017. I have attached the PowerPoint as a Google Slides link, but the presentation was not recorded. The second is a poster, which is also included as a Google Slides link. I found it challenging to try and report all my findings in the limited space of a poster, but in the end I think it turned out nicely! The final format is a technical report, which is included as a Google Docs link. I was given unlimited space to present my findings and methodologies, and I needed it, as the paper is over 30 pages long! Despite this, I still struggled with the order of some sections, as well as with tense. Note that because Google Docs uses a different formatting scheme than Microsoft Word, the technical report will have a few inconsistencies. These include figures having different color schemes and not having borders, sections not matching the Table of Contents, and so on.

presentation

poster

technical report

update: 1/17/18

I have examined the difference between albedo at visible wavelengths and albedo at infrared wavelengths. The first thing I did was repeat the "Extract multi values by points" function, but this time using visible wavelength data. Visible-wavelength albedo data were extracted for the exact same points as the shortwave albedo. Differences between shortwave and visible albedo were then taken in Excel. The results show that wavelength differences in albedo do not fluctuate over time; each location was higher in either visible or shortwave infrared throughout time.

After finding these results, I used the raster calculator in ArcGIS to calculate the difference between shortwave and visible albedo for one dataset; I used an equinox so the entire world would be visible. The results are shown in the map below:

Clearly, areas of land have an albedo value that is higher in shortwave (visible + infrared), while bodies of water and land areas covered by snow and/or ice have an albedo value that is higher at visible wavelengths. This is because water, ice, and snow have higher reflectance in visible wavelengths (0.3-0.7 microns), while vegetation, soil, and rock have higher reflectance in infrared wavelengths (0.7-5.0 microns). For a visual representation of these spectra, see this image (external link).

To conclude, the main reason why Earth's average albedo seemed to be higher in visible wavelengths than shortwave in the Northern Hemisphere's winter was because Antarctica was visible during these months, and this region is dominated by ice. During the Northern Hemisphere's summer months, Antarctica is not visible and average albedo values are higher in shortwave. In addition, land areas covered by snow and ice are higher in visible albedo, which could explain why the average albedo on the Vernal Equinox was higher in visible.

(final) update: 2/12/18

I have repeated the bulk of my analysis, but this time I used shortwave black-sky albedo (BSA) instead of white-sky albedo (WSA). Not surprisingly, I have found the results hold for this type of albedo as well. In some cases, such as the figure below, the graphs look strikingly similar!

This means that both BSA and WSA have the same temporal variations. Because BSA and WSA are only lower and upper limits for the actual albedo, it can also be inferred that the actual albedo undergoes the same variations. Because any more analysis of this dataset will either be redundant or require an extreme amount of more data, I consider this the logical end to this project. I have learned many skills from this project, both mapping/GIS skills and Excel/data management skills! I handled a lot of data and was able to convert this data to information. Overall, I very much enjoyed this project!