An Exciting 4th: Ending the Week with a Bang, not a Fizzle

This fourth week of research brought with it the fourth of July, but I doubt our excitement for Independence Day overshadowed our enthusiasm for the ATom data.

Thanks to Cade (and our new favorite R package: ICARTT from devtools), the ATom data was efficiently converted into an R data frame and made tidy. We decided to investigate eight of the 200+ variables measured by ATom1, resulting in a data frame with 298,120 rows.

The date, latitude, longitude, altitude, relative humidity, dew point, temperature, and ground speed will all help us understand the range of environments in which ATom’s data was collected. These eight variables are just a starter pack for comprehending the more complex relationships we hope to observe later on in the research project.

This gargantuan data set is precisely why we are excited about using TDA. Using standard statistical methods to evaluate this data would be troublesome, and trying to see trends visually via plots would be foolish.

In the later part of this week, we attempted some TDA coding in R with moderate success. Several density cluster tree diagrams were made, but we are still grappling with the ideas that help us interpret these diagrams. This fact did not, however, diminish our satisfaction with our work. The successful creation of any plot at all generally elicits delight in the computer lab.

In addition to exploring the data set, we also spent time learning about various geochemical cycles in the Jacobs book. We read about the cycles of oxygen, nitrogen, and carbon. With this new knowledge, we hope to better grasp the reasons behind the patterns in the ATom data.

In summary, this week has been quite productive and extraordinarily exciting. With the data set finally in hand and the dazzling world of TDA before us, we are eager to explore the implications of the ATom data!