This repository contains the midterm project for the Software Design course at Olin College of Engineering in Needham, MA.
We've noticed that the vast majority of global warming and climate change information in the media centers around either record-breaking disasters or complex scientific studies inaccessible to the average citizen. We decided to visualize weather data over time, not in the Arctic or the stratosphere, but from the same local weather stations that supply your weather app the current temperature when you wake up in the morning.
We decided to dig into data collected by the National Oceanic and Atmospheric Administration (NOAA), more specifically their National Centers for Environmental Information (NCEI). By looking at these historical weather datasets from a wide variety of cities around the United States, we can begin to visualize the increase in global temperatures and ever-changing global climate.
Our hypothesis is that if humans can see the data behind their anecdotal observations of a warming climate, and if these datasets are representative of their own backyards, they will be more inclined to understand the magnitude and urgency of the climate crisis.
This project was developed in a WSL environment running Ubuntu 20.04. Repository setup should be similar when using other operating systems.
This project was developed using the Anaconda Python kernel, and as such our setup instructions will assume Anaconda is being used.
This project utilizes the following list of dependencies:
- Pandas
conda install -c anaconda pandas
- Requests
conda install -c conda-forge requests
- BeautifulSoup4
conda install -c conda-forge beautifulsoup4
- NumPy
conda install -c conda-forge numpy
- MatPlotLib
conda install -c conda-forge matplotlib
The Jupyter notebook included in this project includes pre-made visualizations and analysis. To experience this notebook, no datasets need to be downloaded, as the necessary CSVs are already included in the /data subdirectory of the repository. To view the computational essay, simply run the notebook itself.
If you are interested in exploring the data for yourself, start by exploring the cities listed in WikiMedia Commons at https://commons.wikimedia.org/wiki/Data:Ncei.noaa.gov/weather/{city}.tab.
Then, add entries for your selected cities to the dictionary at constants.py > CITIES, of the format {'City Name For URL', 'city_name_for_file', # of rows to drop}. Next, run scrape_data.py and utilize the helper functions within helpers.py to gather, clean, and plot your data.
Reference function docstrings and comments for more specific instruction.
Andrew Phillips, Brooke Aspen Mague Moss