Develop essential skills to (re)use an ocean of data.

EMBARK CyberTraining Workshops

EMBARK is an NSF-funded, live, hands-on training course for Early Career Ocean Professionals working with large, complex, and long-term open-access ocean data.

Join the community Explore 2027 workshops

The opportunity

Practical pathways through ocean data.

EMBARK is a short-format, live, virtual training course designed to provide hands-on skills for finding, accessing, and using large, complex, and long-term open-access ocean data.

Our open-access course materials will provide practical training in cyberinfrastructure tools and resources from major facilities and data repositories in the ocean sciences.

This training is for the bottlenecks that slow you down and the barriers that prevent you from getting started with incorporating large or complex open data into your ocean data analysis.

What you will learn

Find data from ocean observing systems and data repositories for research scenarios in the ocean sciences.
Evaluate tools to work with those data.
Use coding tools to efficiently access, retrieve, and extract data from remote sources.
Use cloud-based computing for direct access to large-volume data.
Convert data to align with different formats and standards.
Wrangle data for comparison or integration.
Assess the context, quality, and suitability of datasets for use in your research.
Apply a pre-trained AI model to ocean imagery.
Re-purpose code to bring large or complex open data into your ocean data analysis.

Workshop format

This virtual workshop series will cover CI tools, data access strategies, software libraries, and data repository-specific features that enable scientists to efficiently access and (re)use large available datasets

Workshop sessions will include guided small-group breakout activities led by project team members. These activities provide opportunities to work with course materials and connect with other Early Career Ocean Professionals.

Activities will use Google Colab and OOI's JupyterHub platform. Some prior experience with R or Python is expected, but participants will not need to install Python locally.

Session Details

Each cohort includes six live sessions over three weeks. Sessions will likely meet Tuesdays and Thursdays from 2:00 to 4:00 p.m. Eastern Time and include presentations, guided activities, and small-group breakout work.

Session 1

Getting Ready to EMBARK

Learn about the advantages of reusing FAIR data and the variety of data available from BCO-DMO and OOI.

Use Jupyter notebooks, ERDDAP, and the PRIMM model to adapt code recipes.

Session 2

Accessing Reusable Data

Access subsets of large data efficiently with scripts, APIs, and cloud-based computing platforms.

Find, access, and visualize Argo float data using argopy, pandas, and matplotlib.

Session 3

Finding Interdisciplinary Data

Find long-term datasets across disciplinary repositories and use APIs to search for and access them.

Explore EDI and OBIS data and use metadata to assess a dataset for reuse.

Session 4

Working with Large Model Data

Compare model and observational datasets while investigating strategies for accessing and processing large data.

Use xarray and THREDDS to compare ROMS Doppio model data with OOI observations.

Session 5

Classifying Images with AI

Use AI to classify images for further analysis and explore the advantages of GPUs for AI tasks.

Investigate image classification with machine-learning libraries and pre-trained AI models.

Session 6

Extending Your Workflow to Space

Compare satellite and in situ datasets while exploring access to popular satellite products such as SST and chlorophyll-a.

Review satellite provider products and investigate limitations of machine learning for image classification.

Training Opportunities

Cohort 1

January

January 12-28, 2027
Tuesdays and Thursdays

Registration will open on October 15, 2026.

Cohort 2 | Tentative

May-June

May 18-June 3, 2027
Tuesdays and Thursdays

Registration opens approximately 2 months before the cohort begins.

Cohort 3 | Tentative

September

September 7-23, 2027
Tuesdays and Thursdays

Registration opens approximately 2 months before the cohort begins.

An application will be required for each cohort. Participants will be notified approximately two weeks before the cohort begins. Participation is free, with space for 30 participants per cohort.

Stay connected to EMBARK.

Join the community to receive project updates and future registration announcements.

Join the community

Open learning resources

Course materials will live on in the EMBARK JupyterBook.

The JupyterBook will include course materials, activities, and presentations. It will be published at a later date.

JupyterBook coming soon

Frequently asked questions

Participation details

This course is geared toward U.S.-based Early Career Ocean Professionals (ECOPs) who work with ocean data, including advanced undergraduates, graduate students, postdoctoral researchers, and other early-career professionals (faculty, researchers and other professionals in the ocean sciences). This course is best-suited for those who are new to advanced CyberInfrastructure and who encounter roadblocks in their research because they do not have experience or training to handle the volume, variety, velocity, or veracity of data they want to pull into their analyses. Participants must be able to attend all six live sessions of a cohort.

No. Participants who complete the course will receive a certificate from the project.

Participants are expected to be available for 4 hours of synchronous instruction each week, for 3 weeks. Additional time to review materials and work through activities will be needed.

Registration will open on October 15, 2026 for the first cohort. Registration for the other cohorts will open approximately 2 months before each cohort begins.

Sessions are tentatively scheduled to meet on Tuesdays and Thursdays from 2:00 to 4:00 p.m. Eastern Time. Times may be adjusted based on participant interest.

Some prior experience with Python or R is expected. This is not an introductory coding course. Activities will use Google Colab and OOI's JupyterHub platform. Participants will not need to install Python locally.

No. These sessions are designed to include group activities and discussions that are most effective when experienced live.

A JupyterBook with course materials, activities, and presentations will be published at a later date.

Please join the EMBARK community mailing list to receive project updates and future registration announcements. Signup today

People and Partners

The EMBARK Project Team

To learn more about the project team and key presenters, please check out their bios below.

Advisors

NSF Ocean Observatories Initiative

Craig Risien
Jim Edson
Jim Case

BCO-DMO

Adam Shepherd
Danie Kinkade

Environmental Data Initiative

Mark Servilla
Greg Maurer

Rutgers Master's of Operational Oceanography

Alex Lopez
Travis Miles

EMBARK CyberTraining Workshops

Practical pathways through ocean data.

What you will learn

Workshop format