Clemson Research Experiences for Undergraduates

About Me

I am a student at the University of Arkansas studying math and economics. Some of my hobbies are geometry, running, rowing, camping, and baking. A few of my favorite classes have been Abstract Linear Algebra, Probability and Statistics, and Differential Geometry. I am interested in studying Topology at a graduate level and am thrilled to be gaining valuable research experience at Clemson's Data Visualization REU.

Project Description: "Understory light levels in tropical forests are generally less than 5% of full sunlight, and light is considered limiting to tree seedling survival and growth. We employed three quantum light sensors in each of 9 primary forest stands on the island of Dominica in the Lesser Antilles of the eastern Caribbean Sea to quantify understory light levels over the course of 10 months. Each sensor measured the photosynthetically active radiation every 10 seconds during daylight hours. These measurements were then averaged and recorded at 5 minute intervals using a datalogger. The REU student will help to visual these complex data to compare forest stands in terms of overall light levels and seasonal trends."

Week 1

This week I met with Dr. DeWalt to discuss the project and get an idea of the type of data I will be working with. I spent most of the week familiarizing myself with the project and learning to use R.

Week 2

I have been working on 'cleaning' the data in R, i.e. reorganizing the set by site and attempting to identify outliers. We have a complete data set for all nine sites during the month of June, so I plan to create plots with the data from each of the twenty-seven sensors for each day in June. Hopefully, this visualization will help to identify extreme outliers which most likely correspond to sensor misreadings rather than actual spikes in understory light levels.

Week 3

After subsetting the data in R, I used Tableau to create visualizations for each day in June by forest site and PAR sensor. We were able to identify the sensors that gave consistently low readings and compare these to the Gap Light Analyzer (GLA) output. We plan to use R to calculate the average total light per site per day and run a regression to determine the correlation between PAR level and several GLA measures.

Week 4

We found the average PAR level for each site for the month of June and I graphed them using Tableau. I have begun to work more with the GLA data and plan to use other functions of the program in the coming week. Additionally, I began preparing for midterm presentations and the XSEDE conference by writing an abstract and statement of work.

Week 5

I reformatted one of the daily light graphs to be used in my midterm presentation. I spent most of the week preparing for the presentation by creating a powerpoint and determining talking points. I also created additional graphs of the average light levels in June and stacked the data for all nine sites on a single graph to illustrate the consistency in light patterns.

Week 6

I finalized two daily light graphs to be displayed on my poster at the XSEDE conference. I created and edited my poster for the conference and prepared answers to potential questions I might receive at the poster session. I continued to work with the GLA data, but did not include information about hemispherical photography on my poster.

Week 7

This week we were in Atlanta for XSEDE. This conference was my first poster presentation, and altogether it was a positive experience. I was able to answer questions about my research and I received helpful feedback. Overall, I had a very good experience at XSEDE. I met engaging people who were working on interesting research and I learned quite a bit from the talks I attended.

Week 8

I am preparing for the final presentation. I edited my poster based on the feedback I received at XSEDE and added results and conclusions. I am finalizing all of my visualizations to present this week.

Last updated: 7/23/2014