Program

Program at a glance

Detailed program

Please visit again in February 2025 for the detailed program!

Abstracts

Please visit again in February 2025 for the book of abstracts! Click here for the instructions for abstract submission and here for the instructions for the Ron McDowell Student Award, which are now both due on January 24, 2025.

Conference Themes

The overall theme for GeoHab 2025 is Habitat Mapping in a Time of Climate Change. We welcome contributions related to the mapping of marine and coastal habitats, from tropical to polar environments and from the beach to the abyss, using acoustic or optical remote sensing methods, or any other approaches. Planned session topics include (but are not restricted to):

Climate change effects on marine habitats

Coastal and shallow-water habitats

Habitat mapping for marine restoration

Management and communication

New approaches and technologies

Other human impacts on marine habitats

Participatory mapping and citizen science

Polar and sub-polar environments

Shelf and deep-sea habitats

Urban seas systems

Click here for more details on these themes.

Workshop

On Monday, May 12, the traditional pre-conference workshop will be held. Jointly organized by the Circum-Pacific Council and the Florida Geographic Information Office, it will showcase the Florida Seafloor Mapping Initiative, a $100 million effort to map Florida waters to 200 m depth. The workshop will include an introduction of the initiative, presentations by the contractors mandated to collect and process the data, panel discussions of the contractors and the state and federal agencies that will use the data, and a cookbook for other entities that would like to start such project in the future. Do not miss the challenge that will be launched at the end of the workshop! 

Click here for more information!

GeoHab’s Machine Learning Working Group Competition

 The GeoHab Machine Learning Working Group (MLWG) is announcing that the GeoHab 2025 MLWG Compeition is now open! For the competition, participants are invited to build a model to complete a benthic habitat mapping task using a dataset provided. The accuracy of your model predictions will be evaluated using withheld test data, and the top scores will be awarded prizes at the 2025 GeoHab conference!

The competition is being hosted on Kaggle, which is the world’s largest data science community and competition platform. Create a Kaggle account and sign up for the competition to download the data and submit your predictions. The public competition page is your one-stop shop for everything related to the competition, including complete rules, additional information, a temporary leaderboard, and a discussion space to ask questions or exchange ideas with other participants. Use this invitation link to register and participate in the competition.

This competition is a regression task. Participants must predict the mean grain size of seafloor sediment samples given environmental data. The environmental data are provided in both tabular and image (i.e., raster) format, enabling both traditional ML and computer vision approaches. Predictions must be submitted to Kaggle in tabular format (.CSV) per the instructions on the Data page.

Notes: You may register to the Kaggle competition and submit predictions that will appear in the temporary leaderboard, but only submissions made by individuals or teams registered for the GeoHab conference will be considered eligible to receive a prize at the conference. This competition is by invite only. This means that, while anyone can view the public competition page, only individuals with the invitation link (above) may register and compete. Please only share the invitation link with GeoHab members or conference participants (e.g., don’t post it publicly on Reddit). While participants may email the organizers directly with questions, the Kaggle page also has a Discussion board where you can post questions or comments publicly. The board will be monitored, and we also encourage participants to share their notes and prediction maps here.

Stay up to date with the latest GeoHab news!

Subscribe to the GeoHab mailing list.