Tree-mendous Impact

Localized Carbon Footprint Analysis: Driving Effective Environmental Policies

Comment

Inspiration

A recent campus lecture by Prof. Chetan Solanki, the "Solar Man" of India, profoundly impacted our perspective on climate change. His 11-year solar bus journey, a living testament to Gandhian principles of sustainability, underscored the urgency of addressing global warming. Inspired by his unwavering commitment, we are determined to contribute to the fight against climate change.

What it does

Our project helps policymakers by providing a comprehensive carbon footprint calculator that integrates vehicle emissions, industrial data, and tree coverage. Here’s how it supports decision-making:

Emission Estimates: Accurately predicts CO2 emissions from vehicles across different states using a trained ML model.

Industrial Emissions Data: Incorporates CO2 emissions from industries, providing a complete picture of state-wide carbon output.

Tree Coverage Analysis: Uses satellite data to determine how much CO2 is absorbed by trees, showing the effectiveness of natural carbon sinks.

Policy Evaluation: Measures changes in carbon ratings to assess the impact of environmental policies, helping policymakers understand if their actions are effective in reducing emissions.

How we built it

Objective: We are developing a carbon footprint calculator for India that provides insights into vehicle emissions, industrial emissions, and the impact of vegetation on CO2 absorption.

Data Flow and Usage:

Data Collection:

Vehicle Data: We gather CSV files containing vehicle details, such as make, model, engine size, and CO2 emissions for different vehicle types. Industrial Emissions Data: This is collected from CSV files containing CO2 emissions from industries across various states in Canada. Tree Data: Hansen's global forest mapping data is used to assess tree coverage and CO2 absorption capacity. Data Processing:

Vehicle Data: The dataset is preprocessed to handle missing values, encode categorical variables, and feature engineering is performed. Industrial Data: Directly used as-is for emissions analysis without additional ML processing. Tree Data: Processed using Google Earth Engine API to estimate the number of trees and their CO2 filtering capacity. Model Training:

Algorithm: A machine learning model, specifically a Linear Regression algorithm, is trained to predict CO2 emissions based on vehicle features. Training Process: Data Preparation: Features and target variables are prepared from the vehicle dataset. Splitting Data: The data is split into training and testing sets. Model Training: The Linear Regression model is trained on the training set to learn the relationship between vehicle features and CO2 emissions. Model Application:

Prediction: The trained model predicts CO2 emissions for different vehicles based on their features. Total CO2 Calculation: Combined emissions from vehicles and industries in each state are calculated. Tree Impact: The CO2 absorption capacity of trees is assessed and compared against the total CO2 emissions.

Policy Analysis:

Evaluation: The CO2 emissions data and tree filtering capacity are used to evaluate the effectiveness of existing and new policies. Decision Making: Policymakers can use the insights to make informed decisions about environmental policies and their effectiveness based on changes in carbon ratings.

Design: The UI was designed using Figma

Challenges we ran into

Data Accuracy: Ensuring the reliability of vehicle and tree coverage data. Model Performance: Handling inaccuracies in the ML model or its predictions. Regional Variability in Carbon Emissions: Carbon emissions vary widely by region due to differences in vehicle types, fuel use, and industrial activities. Aggregating data nationally or at the city level can obscure these differences, leading toineffective policies

Accomplishments that we're proud of

Here are the key accomplishments to be proud of:

Innovative Solution: Developed a comprehensive carbon footprint calculator for India, integrating vehicle emissions, industrial data, and tree coverage. Effective Use of AI: Trained a robust AI model using Canadian vehicle data to estimate emissions, demonstrating the flexibility and effectiveness of AI in handling diverse datasets. Comprehensive Data Integration: Successfully integrated vehicle emissions data, industrial carbon emissions, and global tree coverage data to provide a holistic view of carbon footprints. Actionable Insights: Provided policymakers with actionable insights on carbon emissions and the effectiveness of policies by tracking changes in carbon ratings. User-Centric Design: Created an intuitive UI with Figma, enhancing user experience and accessibility of the carbon footprint calculator.

What we learned

Here are some key learnings from the project:

Data Handling: Gained experience in managing and processing diverse datasets, including handling missing data and integrating multiple sources of information. AI Adaptability: Learned how to train machine learning models on datasets from different regions (e.g., Canadian data for Indian emissions) and adapt them to local contexts. Model Evaluation: Understood the importance of evaluating and validating models to ensure their accuracy and relevance in real-world scenarios. Data Integration: Developed skills in integrating various types of data (vehicle emissions, industrial emissions, tree coverage) to create a comprehensive analysis. Policy Impact: Realized how data-driven insights can influence policy decisions and how tracking changes over time can help assess the effectiveness of environmental policies. UI Design: Enhanced skills in creating user-friendly interfaces that make complex data accessible and actionable for end-users.

What's next for Tree-mendous Impact

We chose to build this project to address the pressing need for effective carbon footprint management. By integrating both past and future datasets, our system provides a dynamic view of how carbon emissions are changing over time. This comprehensive approach allows policymakers to evaluate the effectiveness of their environmental policies and make informed decisions based on real-time data. The ability to track changes in carbon footprints enables us to assess the impact of different strategies and identify areas for improvement. As we incorporate more datasets and refine our models, our tool will offer deeper insights, helping to drive advancements in carbon management and support more effective climate action initiatives.

Share this project:

Updates