Inspiration
A member of our team had been exposed to new sustainable farming methods during his internship last summer in Singapore. He rallied his teammates behind this newfound inspiration, and once we saw the Stanford Ecopreneurship challenge, we knew we had to make this! We wanted to gamify learning about the world's carbon emissions, to spread knowledge about the topic and encourage better mindfulness!
What it does
CARBONLE is a geographical Wordle-like guessing game where you have up to 6 guesses to identify the 'target country'. Our application leverages real-world carbon emissions data and displays them with Treemaps (like the square graphs that Stock changes are displayed on!).
How It Works:
- A random target country is chosen at the start.
- Enter your guesses in the input field. Each guess reveals hints about distance and accuracy, guiding you closer to the target.
- Treemaps are displayed for both the target country (showing sector-based emissions) and your most recent guess (to compare sector emissions).
- If you run out of guesses or guess the country correctly, a final result screen appears.
- After the game ends, an OpenAI-based summary is displayed with environmental information about the target country.
How we built it
We took many steps to complete this application. We built our frontend from scratch using create-react-app, utilizing graph libraries like Rechart and API libraries like OpenAI.
We built and gathered our own datasets. We wrote scripts that utilized APIs and data wrangling methods to obtain Sector and Sub-Sector Carbon Emission data per country. The APIs we used were from Climate Trace and Climate Watch.
Challenges we ran into
We had two main challenges. The first main challenge was wrangling the data into a usable format for our Rechart Treemaps due to a lack of proper API documentations and limited API endpoints to obtain data. Our second challenge was facing the limitations of the Rechart Treemap library, amongst other alternatives. Each Treemap library faced coloring and border issues when it came to nesting Treemaps within a hierarchical Treemap. Thus, we adapted and overcame these obstacles through clever API endpoint scraping concatenation and clever paired UI displays of Treemaps.
Accomplishments that we're proud of
We strongly believe that the datasets that we created throughout this hackathon project are the most accurate and comprehensive carbon emission datasets; we believe we have improved the holistic-ness and usability of the Climate Trace dataset specifically. We also all learned a lot about climate and sustainability throughout this project, and we all feel better educated!
What we learned
China is a HUGEEE contributor to carbon emissions! We also learned that there are some countries out there with negative carbon emissions based on their LUCF (Land use change, and forestry). We also learned that creating a gamified version of a climate game genuinely increases our desires to learn more about this topic even beyond the scope of this hackathon.
What's next for CARBONLE
We spent hours trying to fix the colors on our Treemaps for sub-sectors. We definitely want to finally find a solution to this, so that we can display a more clear Sub-Sector UI. We would also like to add more fields so that users can filter between different types of emissions!
Log in or sign up for Devpost to join the conversation.