Project Info
Data-driven and visualization tools for improving the quality of oceanographic data
Bia Villas Boas
villasboas@mines.edu
Project Goals and Description:
The oceans are the largest carbon sink on the planet, regulating global climate by absorbing vast amounts of CO2 from the atmosphere and storing heat. Therefore, accurate oceanographic data is crucial to understanding the complex interactions between ocean and climate systems. However, the quality control of oceanographic data is still a major challenge. Traditional quality control methods are often subjective and time-consuming, and human errors can lead to data inaccuracies.
This undergraduate research project aims to develop and apply data-driven methods to automate and improve the quality control of oceanographic data. The student will explore conventional statistical analysis and machine learning approaches to identify and correct errors in the data. An additional goal of the project is to develop web-based visualization tools to facilitate the exploration and interpretation of the data in collaboration with an international team of expert oceanographers.
This project offers an excellent opportunity for students to enhance their computational and programming skills while contributing to a real-world problem of great societal relevance.
More Information:
Grand Challenge: Not applicable.
- Website of the International Quality-controlled Ocean Database (IQuOD)
- Castelao, Guilherme P. "A framework to quality control oceanographic data." Journal of Open Source Software 5, no. 48 (2020): 2063.
- Castelão, Guilherme P. "A machine learning approach to quality control oceanographic data." Computers & Geosciences 155 (2021): 104803.
Primary Contacts:
Bia Villas Boas, villasboas@mines.edu
Student Preparation
Qualifications
To succeed in this project, the student should have experience with basic statistics and probability, basic knowledge of Python, and be motivated to learn version control, collaborative software development, and project management through GitHub.
TIME COMMITMENT (HRS/WK)
4
SKILLS/TECHNIQUES GAINED
- Gain knowledge of data analysis and classification techniques using machine learning.
- Learn and practice developing Python packages following industry best practices such as CI/CD.
- Develop an understanding of basic physical oceanography
- Build intuition around data visualization systems
- Improve collaboration and communication skills by working with an international and diverse team of experts.
MENTORING PLAN
This project will be carried out in close collaboration with scientists at Scripps Institution of Oceanography, UC San Diego. The faculty mentor (Dr. Villas Bôas) and the students will meet with the external collaborators weekly over Zoom. Additionally, the student will be welcome to join weekly Mines Oceanography team meetings, which will provide exposure to a research environment and an opportunity to connect with other students and researchers on the team. Depending on student interests, goals, and progress, this project may also lead to scientific publications and presentations at scientific conferences.
PREFERRED STUDENT STATUS
Sophomore
Junior
Senior