2024 Summer Internship at PolarWatch
Internship (Satellite Data Scientist) Details
- 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/
- Communications
- Daily checkin via Google workspace Today I will work on ..., Tomorrow I will work on .. I had issues on ..
- 30 min 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
Project Milestone and activities
Milestone 1 ( 2 wks: 06.03 - 06.14)
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
Milestone 2 (3 wks: 06.17 - 07.05)
Objective
you will learn about one sea ice data from PolarWatch(sea ice concentration, thickness, IMS, etc.)
Activities
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
Milestone 3 (3 wks: 07.08 - 07.26)
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)
Milestone 4a (2 wks: 07.29 - 08.09)
Objective
You will learn to model sea ice prediction using machine learning (deep learning) apporach. we will replicate the
model introduced in the paper (https://s3.us-east-1.amazonaws.com/climate-change-ai/papers/icml2021/50/paper.pdf)
Milestone 4b (1 wks: 08.10 - 08.16)
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
Resources