Using Scripps Institution of Oceanography’s Global Observatories to Study Atmospheric CO2

This is an updated post from my old site that I decided to rewrite. This first project is a great way to learn about climate science and get an introduction into using climate datasets.

The data for this project can be found here: https://scrippsco2.ucsd.edu/data/atmospheric_co2/sampling_stations.html

I wrote my code for the data analysis on Jupyter Notebook which I will attach at the end of the this post, but enough blabbering onto the project!

---

The first step to any project like this is to process the data so that it can be used. This step involves removing any outliers that don’t belong (maybe because of a data entry error, or any number of other reasons), or removing metadata, or simply turning the data into usable data types. In this case I had to do all three. First thing I did was skip over the metadata that talks about credit and sampling techniques, and while important, does not help our data analysis. Second step was that I noticed some of the data points were totally nonsensical. For example the first two months of the Mauna Loa data had CO₂ concentrations that were simply impossible (-99.99 CO₂ ppm). Then finally, I turned all of the data types from objects to numbers.

Some helpful functions to start the preprocessing are the following:

print(df.dtypes):
# This will give you a list of all the types that are in your dataframe (assuming you're using pandas). This can honestly save you so much frustration. Sometimes when you are looking at a file the inputs may look like numbers but the way it was actually recorded can be a totally different input type (often object)

df.head():
# This gives you a little preview of your data. It is often helpful just to get a sense of what types your data are, or just get a grasp on what you're dealing with.

df.columns = ["name1", "name2", ... , "nameX"]:
# I often like to rename my columns because often names are not very helpful, or sometimes they just have really weird spacing and make it hard to call.

df.info():
# This gives a quick little summary about your data and can often give insights about the number of outliers in your data!

Moving on to our actual analysis lets start plotting our data. For this example we are looking at the Mauna Loa Observatory which is located in Hawaii. This dataset recorded the atmospheric CO₂ in parts per million (ppm) on site. The dataset from the Mauna Loa Observatory has been gathering data since 1958 and continues to publish their results to this day. I plotted the data on a graph to look at how the atmospheric CO₂ has changed which you can see in fig. 1

The most striking thing we can see from this graph is obviously the steep increase in atmospheric CO₂ over the years; however, I was also quite intrigued by the very consistent fluctuations we see in the data. I decided to also plot the monthly atmospheric CO₂ values on their own (fig. 2) and in comparison to each other (fig. 3). After realizing that fig. 2 was not very helpful I decided to plot the different months compared to each other as can be seen in fig. 4.

I thought this was a really cool realization! After looking into it I learned that the reason we see a decrease in atmospheric CO₂ is because of plants!¹ It is also important to note that these observations are being taken in Hawaii which is in the northern hemisphere. As we head into the spring (April – May) more plants come out and as a result of photosynthesis they capture CO₂ and convert it into oxygen and sugar. Which is why after May we start to see if steadily fall until October. By October a lot of the leaves have fallen from trees and many plants have started to decay releasing the CO₂ they captured in the spring and summer months. Over the course of winter there are no plants to absorb the atmospheric CO₂ and with increased energy consumption to heat homes² (in the northern hemisphere) we see a steep increase in atmospheric CO₂ until we reach May and the cycle starts all over again.

Unfortunately, as the data shows, this isn’t a cycle; but rather a spiral. As we saw in figs 1, 2, and 3, we have seen an increase in atmospheric CO₂ which is a well known greenhouse gas. It is issues like these that we face, and it is up to us to decide how to move forward.

---

In hopes to not make this post too long I will make another post that looks at and compares the other Scripps observatories so we can get a better idea of where in the world is being affected the most.

Resources:

Data:
https://scrippsco2.ucsd.edu/data/atmospheric_co2/sampling_stations.html

Research Citations:
C. D. Keeling, S. C. Piper, R. B. Bacastow, M. Wahlen, T. P. Whorf, M. Heimann, and H. A. Meijer, Exchanges of atmospheric CO2 and 13CO2 with the terrestrial biosphere and oceans from 1978 to 2000. I. Global aspects, SIO Reference Series, No. 01-06, Scripps Institution of Oceanography, San Diego, 88 pages, 2001.

1. https://www.climate.gov/news-features/understanding-climate/climate-change-atmospheric-carbon-dioxide ↩︎
2. https://www.eia.gov/todayinenergy/detail.php?id=42915 ↩︎