- Intern: Tien Ly from San Jose State University
- Mentor: Sunny Hospital from PolarWatch
- Start Date: June 3, 2024 (Mon)
- End Date: August 16, 2024 (Fri)
- Project Repository: https://github.com/polarwatch/internship24/
- Communication:
- Daily check-in via Google Workspace Today I will work on ..., Tomorrow I will work on ..., I had issues on ..
- 30 minute video check-in on Mondays and Fridays to talk about progress, issues, adjustments, etc.
- Mondays to focus on technical work discussion
- Fridays to focus on culture and constructive criticism/advice/discussion for mentor and student
- Bi-weekly PolarWatch team meeting to meet the team, updates, etc.
- Project Management : GitHub issues
- NOAA Coastwatch seminar series
Objective You will learn how to use ERDDAP data server, satellite data, python package xarray.
Activities
- Satellite course indicated in resource section
- Satellite 101
- Sea Surface Temperature,
- Sea Ice,
- Sea Surface Height, Winds, Salinity,
- Tools & Strategy (excl R and ArcGIS)
- Installation of python and required packages
- conda, python, and required packages provided in requirements.yaml from mentor
- Complete some python tutorials from coastwatch training github
- Setting up a Python environment
- Tutorial-1
- Tutorial-2
- map-data-with-different-projections
- calculate-seaice-extent
Deliverables
- Set up conda environment using environment.yml file
- GitHub - use of
git clone,git pull,git push,git branch [your_branch], submission of pull request - Add Jupyter notebooks to the
notebooks/directory (showing python code used for the satellite course) - Add course summary (summary of what you’ve learn) to the
reports/milestone1/directory - Add Activity report (briefly describing what you have done) to the
reports/milestone1/directory
Objective You will learn about one sea ice data from PolarWatch(sea ice concentration, thickness, IMS, etc.).
Activities
- Learn Polarwatch catalog
- Learn about how the data of your choice is developed
- Learn to read metadata and describe the data product based on the metadata
- Literature review - find 3-5 journals on how the data are used in research or application and mentor/intern will select 1 article to read
- Create Python notebook to generate statistical summary and visualization of the data from PolarWatch
Deliverables
- Description of the data product
- Summary of the article
- Jupyter notebook with data summary and visualization
- Activity report (briefly describing what you have done)
Note If Tien finds an interesting project and the internship schedule allows, we will try to incorporate modeling component.
Objective You will learn to compute climatology and visulize data on a polar projected map. Additional Activity: buidling predictive model using sea ice and machine learning
Activities
- Learn climatology of sea ice data
- Complete tutorials to learn how to comptue climatology climatology
- Download data and compute climatology of your sea ice data
- Draft Jupyter notebook with explanation
- Draft a presentation and present data summary to the team
Deliverables
- Jupyter notebook
- Data summary report
- Activity report (briefly describing what you have done)
Objective You will learn to model sea ice prediction using machine learning (deep learning) apporach. We will replicate the model introduced in the paper
- Read journal article: https://s3.us-east-1.amazonaws.com/climate-change-ai/papers/icml2021/50/paper.pdf
- Create python notebook to build the predictive model using sea ice data from PolarWatch and LSTM algorithm
- Git repo: https://github.com/big-data-lab-umbc/sea-ice-prediction/tree/main/climate-change-ai-workshop
Objective You will learn to draft all your work in Quarto and publish in github.io https://quarto.org/docs/gallery/
Activities
- Design and develop a project website
- Set up quarto/python
- Draft a method page for your data and analysis
- Deploy project on github.io
- Give the project presentation to the team
Deliverables
- Deployment of the webpage
- Presentation of the project
- Final report