Carbon chat| What Is Our Carbon Footprint?
Awards & Nominations
Carbon chat has received the following awards and nominations. Way to go!
What Is Our Carbon Footprint?
Making CO2 Data More Accessible to the Masses
Summary
Lets making CO2 data available to anyone that owns a mobile device!Our project is designed as an iOS app which can triggered pointers on a map for a user to see CO2 emissions. We created a polynomial regression model which utilizes OCO-2 XCO2 observations to analyze carbon dioxide in the atmosphere based on specific locations and output future CO2 emissions. Currently, our project's details are based around Waitangi, New Zealand due to the high volume of data in that area but our algorithm can be applied to states around the globe to persuade local policy makers to understand how CO2 levels have been increasing and have the potential to increase exponentially, as shown in our model.
How We Addressed This Challenge
Our project is an iOS app which was integrated with a Heroku backend site which could connect to our Python machine learning models. Through the iOS app frontend features are provided and a map is displayed on the screen with multiple points displaying information on CO2 levels in a specific area. To retrieve CO2 emissions for the area in real time (as predicted by our polynomial regression model), the app performs an API call to the heroku site and retrieves this information based on the current date. Then, our polynomial regression model takes this time input and returns the predicted CO2 levels to the heroku site to be displayed to the user on the app instantaneously. We hoped to achieve a user friendly app that can possibly influence future policy makers to realize the need to promote environmental change in local areas.
How We Developed This Project
Our team chose this challenge because we wanted to work with a machine learning model to create future predictions based on data. Seeing the OCO-2 XCO2 Observations as a clear opportunity, we took on this project for helping the community by analyzing CO2 data and displaying it in a user friendly way. We utilize Swift, Python, Heroku, and machine learning libraries in Python like numpy, matplotlin, and pandas to perform these tasks. We had multiple issues with retrieving data from NASA, being unable to open .nc4 files and trying to understand how to analyze such high volumes of data. However, we were able to solve this problem as a team and create functions to perform effective data analysis.
How We Used Space Agency Data in This Project
We used the OCO-2 XCO2 Observations dataset to analyze multiple .nc4 files. We then combined all of these files (including thousands) into a single .csv file to be analyzed by our machine learning model. After separating it by location, we chose New Zealand and analyzed the data accordingly.
Project Demo
https://docs.google.com/presentation/d/1g32GarTieCUAeKttDZFoZQiqNQ2yOq567wwRsJufmZA/edit?usp=sharing
Project Code
Data & Resources
https://disc.gsfc.nasa.gov/datasets/OCO2_L2_Lite_FP_9r/summary?keywords=OCO-2