Core Data Analysis-2 Class assignment
Notebook/Homework 12
Due Wednesday March 20, 2024
This is a true story. A few years ago I was working with a graduate student from Lamont/Columbia. We were helping analyze soil for lead at a farmer’s marker in Williamsburg. She wanted to expand her work and to look at lead in backyards in Brooklyn. We flyered and used facebook to find people willing to let us sample their backyard. We went into people’s backyards and collected soil cores. We had not idea what we were going to find. We only had data from studies published in other parts of the city. We told the graduate student to present what they found and what they think. For this homework you are that graduate student! You are going to present on one of the cores from the study. I will email everyone the core to use.
This week we are going to develop a hypothesis to test on core data. When you come in for class we are going to sit down and read and discuss the Chillrud paper. Read it so if you have questions we can discuss them. The more you ask the better you will be.
Files That may be helpful
- All data and notebooks are in the core-data folder on GitHub.
- Read the Chillrud Paper (Chillrud-es9807892)
- Listen to the Podcast. Here is the itunes Link. Choose Chillrud_Lead
- Read the Earth Institute Blog post about Sampling in Brooklyn.
- notebook (pdf=plotcoredata-Brooklyn) for background
- The paper that Franziska Landes (pdf=Landes-1-s2.0-S0048969723040305-main-2) published and the data from the paper (excel=Landes_STOTEN or on github).
- Data from Table 2 in Chillrud
- Notebook with the simple answer (pdf=plotcoredata_Landes_answer-Brooklyn-Inventory)for adding axes
You are going to read a paper (Chillrud-es9807892) by Chillrud et al., 1999 titled Twentieth Century Atmospheric Metal fluxes into Central Park Lake, New York City. They have some interesting conclusions. This was all we had before we started to collect cores (there are a few other studies but this is the best for understanding processes). As part of a new study we have collected soil samples and Cores from Northern Brooklyn. You will analyze one core from the Landes paper. The cores are highly contaminated. We want to determine the source of that lead. Your goal is to develop a hypothesis and test it. (One plot of Pb v versus depth will not get you far and at best will get you a 50 for the assignment). Plus you will need some writing to explain your thinking. So use Markdown to explain the process and to create figure captions.
Remember that analyzing and thinking about data is hard and takes time. You need to think of a strategy for analyzing the data and what you want to accomplish. What is your hypothesis? What are asking? You know how to do a lot of different graphs and analyses, figure out what you want to apply. You can make plots versus depth. You can correlate parameters. You can do math to calculate numbers. You can look at a lot of data in many different ways. Use your knowledge to test your hypotheses. Remember you can iterate on your question/hypothesis as you analyze the data. It is a two-way street. Make some plots go back re-read. Think some more. Hone you hypothesis. Look at all the graphs you have learned to make. Think about what you can do with the data. Chillrud didn’t have course and fine grain data. Can that help you think about sources?
I found this class to be quite difficult. While reading the paper and listening to the podcast about the subject matter was straight forward and interesting, I didn’t fully understand what we were expected to do for this class at first. However, as we had two classes to work on this assignment, I eventually figured it out. Though the instructions were not the most straight forward, I think this assignment prepared us for the final project in that we had to formulate a hypothesis and figure out how to manipulate our data in order to best communicate it. I think that spending 15-20 minutes at the beginning very clearly explaining what the goal and purpose of the lesson were would be helpful.
This assignment was interesting and a very useful one. However, I found it a little difficult to understand we were supposed to do at the begining of the class. I found the podcast and other background articles extremeley useful
This assignment was definitely not straight forward. However, I think it prepared us well for the final project! It allowed us to view data and try to analyze it on our own. I think it was super helpful to listen to the podcasts and read the article before class. Maybe a longer discussion in class of the article would be helpful for background information. However, I still thought that this assignment would still be straight forward upon understanding the scientific aspect of the data beforehand.
I will work on more discussion and maybe background
I really like how this class allowed us to come up with and test our own hypothesis, but I wish that we could have had a little more choice in which paper to use since we all ended up doing relatively similar tests. Perhaps in the next class you could allow students to choose from a couple different papers to base their hypotheses on.
I found this assignment to be super difficult, probably due to lack of structure given on how to use the strategy applied in the research paper to our own dataset. I particularly spent quite a bit of time trying to understand the radionuclide dating procedure from Chillrud, to then realize that that sort of analysis was not available to us with the Brooklyn core data. I think hosting a class discussion to draw out the proposed research methodologies of those students that come from environmental science backgrounds could have been helpful. It was really cool that we were trying to do real science, but I ultimately felt like I was only able to achieve very blunt analyses and to draw broad or inaccurate conclusions from them. Learning how to plot twiny and twinx axes was very useful for later down the road!
I really liked this assignment and found plotting different parameters with multiple axes super useful for my poster project. I think that the podcast was really helpful for figuring out what elements to look at and where to look for relationships so I highly recommend emphasizing its usefulness in the future. The only thing that I don’t think I could have gotten without help was inventories, which could maybe be given a few more hints on.
Echoing many of the previous comments, this was one more the more difficult and time consuming assignments, but also an extremely valuable one. This was the first assignment that really showed me how I could potentially use python in my own environmental/general research outside of a classroom setting. While I appreciated the practice of using python to analyze a hypothesis I had created, I wish there was a little more structure in what was expected as the final project. While Brian did state that 1 plot wasn’t enough, I found myself second-guessing whether I needed to add excess plots and analysis just to fill up space or cover my bases because I didn’t know how he was going to grade this notebook. Also, I wish we’d spent a little more structured time talking about the conclusions of this paper in class (especially for people who didn’t have in depth environmental science backgrounds).
While this assignment was one of the harder ones, it was definitely one of the more creative and interesting assignments. I really enjoyed how we could frame our own hypothesis, and use data to either support or reject it. I also found the topic incredibly fascinating; after doing this assignment, I felt like I had accomplished a lot!
While I enjoyed the freedom of this class, I found “making sense” of the data challenging. To create meaningful plots for this assignment, it’s very important to have a background in data analysis in order to come up with relevant figures. In the future, it may be helpful to have an additional paper or two to help guide students in a meaningful direction.
Totally agree with this comment. A lot of other commenters have talked about how they found this difficult because they don’t have specific environmental science background. I do have that background, and I thought this assignment stood out to me because it reminded me of my thesis work, where it was up to us to test the associations between different variables. However, when you are working on a thesis project, or doing research, you usually do a big review of literature first to get very familiar with the topic, and obviously we don’t have time for that. I’d suggest instead of using lead toxicity, which is a topic that many people don’t have expertise in, you could do a similar assignment with something more intuitive, even using the other data we talked about in other classes, like sea level rise, air quality, etc. That way, people who aren’t in the major wouldn’t be so immediately lost.
While this was one of the more time consuming homework assignments, it was also one of my favorites. I really liked how this homework was structured as an experiment in a way, by having to come up with a question and then use the data to figure it out. I would recommend doing that for more assignments because it made it feel more relevant and applicable to what we might do in the future. It was also really cool figuring out how to make graphs with different axes for different elements as that is harder to do in excel i think. overall really enjoyed this assignment and it was one I actually felt kind of proud of after finishing.
This class and activity was incredibly valuable and a great introduction to “real” big data! While it was liberating to take this in whichever direction we saw fit, I feel it would have been beneficial to discuss ideas and collaborate as a class. I would have loved to understand the ways in which my classmates chose to analyze the data and their findings.
I really liked this class; it definitely stuck out to me for a lot of the reasons already commented here. I felt like I was really getting a good grasp of graphing and seeing it used to analyze real data with real imagery was fun. The homework also matched the class well.
(Also, a somewhat picky and more general comment, rather than this class particularly — it would be nice if you went back into some of the notebooks and edblog posts/homework assignments and edited some of the spelling and capitalization mistakes. This isn’t a big deal but was something that kind of bothered me throughout the semester when occasionally trying to decipher things.)
I loved this assignment because, like an earlier commenter said, it really ‘took the training wheels off.’ I enjoyed the flexibility of being able to interpret the variables that interested me, and it was also nice to bring environmental literature into the mix. Further, it built off of some topics from my wetlands and climate change class that I took last fall, which was a nice connection to note. As for the content of the lecture itself, I didn’t really find any major problems doing the work. Perhaps some more general background information on relationships between basic elements, weather patterns, human sources, etc would be beneficial to those doing this assignment without much environmental background. Perhaps a short definitions sheet describing each column, saying something like what comes from where, why it matters, relationships, etc.
I enjoyed this assignment and thought it was very valuable to approach the data with a hypothesis. However, I wish we had spent a little bit more time discussing the Chillrud et al. paper because I felt confused at first about what types of questions we should be asking.
I liked this assignment too. The plots were visually effective, esp. when coupled with the photo of the core sample. Being able to plot depth and concentration in one visualization is powerful. I had concerns about the different units of the elements being plotted on the same image, but, overall, seeing parallel fluctuations in concentrations is an effective way to understand interrelationships.
I really liked this assignment! It was extremely challenging for me because it was integrated with real research. Not only did we have to plot in python, but we had to read an article and try to extract background information from Chillrud et al… That part was really exciting for me! However, it was difficult to figure out what to plot… I remember feeling a little overwhelmed because for the first time we not only had to interpret information and decide what to plot on our own.
This assignment was the hardest for me. I think it was because I do not come from an environmental science or any science background, so reading “Twentieth Century Atmospheric Metal fluxes into Central Park Lake, New York City” was confusing for me. Consequently, it made it difficult for me to know if I was analyzing the data correctly.
But I liked the fact that we “took the training wells off” on this homework assignment, and it was up to us to figure out how to analyze the data ourselves -even though I am not sure if my analysis was correct.
This assignment is very exciting for me – when I did my thesis research, I looked at sediment cores to determine the effect of oysters on nitrogen in sediments and coastal estuaries. I analyzed many a sediment core – being able to represent that data with the use of python would have been enormously beneficial. However, at the time I did not know python. Moving forward, the ability to use python to directly interpret core data (as well as other data) will help me with projects.