FlowCast: Real-Time Water Monitoring and Prediction

Our application leverages machine learning and real-time NOAA data to forecast essential water quality metrics like oxygen levels and pH with interactive data visualizations.

Comment

Inspiration

FlowCast was inspired by the increasing importance of water quality monitoring in our environment, particularly in areas like Biscayne Bay and Haulover Beach. With the growing concerns over pollution and ecosystem health, we aimed to develop a tool that provides real-time data and predictions to help stakeholders make informed decisions about water safety.

What it does

FlowCast is an innovative web application that visualizes and analyzes water quality data in real-time. Users can explore various interactive features, such as scatter plots, maps, and line charts, to gain insights into the health of water bodies. The app also allows users to upload their own data, view raw data tables, and stay informed with updates through a user-friendly interface.

How we built it

We developed FlowCast using Streamlit for web application development, Python for data processing, machine learning, and storing historical data. We integrated the NOAA API to fetch real-time water quality data, enabling the app to provide current and relevant information. Our team utilized various libraries such as Pandas for data manipulation, Matplotlib and Plotly for visualization, and Scikit-learn for machine learning model training.

Challenges we ran into

During development, we faced several challenges, including:

  • Integrating real-time data from the NOAA API and ensuring its accuracy.
  • Handling missing features in the prediction model, which initially led to errors.
  • Designing an intuitive user interface that balances functionality and ease of use.

Accomplishments that we're proud of

We successfully created a fully functional web application that visualizes water quality data and provides predictions based on machine learning models. Our team also developed a robust user engagement feature, allowing users to upload their own data and receive updates. Additionally, we are proud of the collaborative spirit and problem-solving skills we demonstrated throughout the project.

What we learned

Throughout the development of FlowCast, we learned valuable lessons in teamwork, project management, and technical skills, including:

  • The importance of data validation and cleaning in machine learning applications.
  • How to effectively integrate APIs and utilize cloud databases.
  • Enhanced our proficiency in data visualization techniques and tools.

What's next for FlowCast: Real-Time Water Monitoring and Prediction

Moving forward, we plan to enhance FlowCast by:

  • Adding more data sources for comprehensive water quality monitoring.
  • Implementing advanced machine learning algorithms to improve prediction accuracy.
  • Expanding user engagement features, such as community feedback and reporting tools.
  • Exploring partnerships with local environmental organizations to promote the app and its usage in water quality advocacy.

Built With

Share this project:

Updates