Final Post - May 2022
Thanks to Senior Projects, I have had the opportunity to focus all of my academic attention on my MARC project for the last three weeks. Despite having one week taken away by my case of COVID, it was an extremely productive period for my project. For starters, I finally was able to extract the NDVI time series that my mentors needed. The key was switching from the python API to the JavaScript API. I had originally chosen to use the python API because I already had a solid baseline from my previous tasks, however Google Earth Engine is originally built for JavaScript, so using that proved to be much more intuitive. After completing this task, I then went on to make my slide show for the presentation. The presentation was unfortunately on zoom, but it was fine. After the presentation I but together my paper, which was an interesting process given my project to a little bit of a different style. For my "Results" section I talked about the before and after of the data I created, and for the "Discussion" section I talked about what is the next step in the project. Now that I am officially done with the MARC program, I am a little bittersweet because I have had a really good time. This was obviously a really unique opportunity and it has opened my eyes to the amazing experiences that come with being involved in the research community. I plan on remaining in contact with my mentors and if they have more tasks for me I'd love to do them, and if the opportunity ever opens up in college to do research alongside one of my professors, I will jump at the opportunity!
February Update 2022
This month I continued to work on the NDVI download. Now that I knew the specific satellite, time range, and the locations that I need to download data for, the only thing left to do was to actually download the data. I found what I thought was a good guide to follow, but after spending many tedious hours typing up the 200 coordinates that I need to download data from, the code ended up not working because the program was not recognizing any of the NASA satellites. I think the problem was that my directories are messed up on my laptop, so GEE api and some of my python add ons are not able to work together. I tried deleting and redownloading GEE api to the same directory as the other stuff I need in the program, but I wasn't able to do it. I think the best thing for me to do is to try the same code on my desktop, unfortunately though, it is not possible to transfer the code over (to my knowledge), so I need to re-write it. In addition to the coding work I did, I also spent some time getting an outline ready for my intro paragraph of my paper. I hope to have a draft done by the end of next month.
January Update 2022
I spent the bulk of this month working on
December Update 2021
This month I began working on learning how to use Google Earth Engine in Jupyter Notebook. Many of the Google Earth Engine tutorials that are made available by Google use their JavaScript console that is directly built into GEE, so finding tutorials has been a little less straightforward than I would have liked. That being said, they definitely exist; I have had some success on many of the same forums that I used in the past, with a couple of developer resource and blogs, and youtube videos.
October Update
2021
October has been a little bit of a slower month. The work I did combining and organizing data was successful and now my mentors are going to have me working with Google Earth Engine. My account with GEE has been approved and I have downloaded a python extension to work with GEE, so I am ready to begin as soon as I am given a specific assignment.
Late September Update 2021
Up to this point, September has been a productive month. I have continued to add to and improve my website and have also been working on the next task my mentors have given me. This task was to do something similar to what I had done with the previous data they had given me (that being to combine and reorganize data), but this time, the data spans all the way from 1920 meaning between all the stations there are almost 1.4 million rows of data. The data is more simple (Daily Temperature Max/Min/Median), but since excel only allows a little over 1 million rows, the script I had previously written in Jupyter Notebook cannot be directly applied. This means that I have needed to divide the data into two groups, which complicates things.
The second year of the MARC program has arrived and It definitely feels as though my project is definitely not halfway done. I definitely have a lot to do, but If I stay consistent and productive throughout the year I will have plenty of time to bring my project together. As of right now, My two main focuses are working quickly and intentionally with my mentors and catching up/remaining caught up with classwork, blogs, etc. Up to this point, my mentors have been giving me work to do every few weeks, so when I am given that work, returning it ASAP is the best way to move my project forward. In the gaps between this work, I plan to focus significant time on my website (more blogs, pictures, etc) and classwork.
This summer played a key role in the advancement of my MARC project. To start the summer, I was focused primarily on finishing up and organizing any not yet completed work from the previous semester. The next few weeks of June were pretty intense. I was working 4, 5, 6, and even 7 hours in a day on figuring out the various training tasks my mentors had asked me to do in Jupyter Notebook. This work was challenging in the fact that the work itself would not necessarily have taken very long if was very experienced, but since I wasn't, working through error codes and trying to figure out what they meant was, as mentioned, time consuming. The middle of the summer was much less productive than the start. I left for two weeks and was completely off the grid, and when I returned, I found that my mentor had injured himself and was in and out of surgery for the next month. In August, though, we got back on track and I was tasked with the first non-training assignment. This task was to combine and re-organize rain data from 84 Israeli weather stations over the past 15 years. in this case, there were multiple steps in the process and I had to go from excel, to python, back to excel, and then back to python again. But, at the end of the day, when I successfully completed the task, the feeling was like no other.
I am currently in the early stages of working with my mentors and am learning about various important aspects of what I will eventually be doing. This training has included learning how to plot and organize satellite data in the form of an XSL, and also how to find various important statistics such as RMSE (Root Mean Square Error), bias, and correlation in Jupyter Notebook (Python). The next things for me to learn is how to download the satellite data from a larger database, and once I am able to do this I should be ready to start working. Me and my mentors are still figuring out exactly what I will be doing, but I imagine I will start with some trials to build up trust, and then hopefully get to working on the official project.