Principal Investigator:
Dr. Art Trembanis
University of Delaware
Project Launch:
October 2012
Project Website:
Case Study - SubseaObservers
Project Background

SubseaObservers (SO) is an online citizen science activity that enables members of the general public to help the US government manage the mid-Atlantic scallop fishery. As opposed to traditional dredge-based surveying techniques, SO provides a means of participating in an innovative new Autonomous Underwater Vehicle (AUV) based approach that takes photos of the sea floor. Once the photos have been collected, scallops can be counted and sized without disturbing them, other marine wildlife, or their habitats.

Each year, between 15,000 and 25,000 metric tons of sea scallop meat are fished from mid-Atlantic waters. In 2010 alone, sea scallop landings accounted for $450 million dollars. Managing the fishery is important to ensure that it is not over-harvested while continuing to meet demand for product. By analyzing SO's underwater images, participants play a direct role in ensuring that surveys are as accurate and comprehensive as possible, which provides information that is critical to establishing fair and environmentally sensitive catch limits. In the process, participants have an opportunity to learn about conservation, the scallop fishery, and how advanced robotics in conjunction with citizen science can help reduce the environmental impact of commercial operations.

The Dataset

Researchers at the University of Delaware have teamed up with Phoel Associates and the fishing vessel Christian and Alexa to develop a new method for surveying the Mid-Atlantic Bight sea scallop population using photographs of the sea floor. As opposed to traditional dredge-based techniques, the team employs the Gavia Scientific AUV platform for survey runs, and their AUV is equipped with side-scan sonar, bathymetric sonar, a digital video camera, and a suite of sensors for measuring dissolved oxygen, turbidity, temperature, density and salinity. Using an AUV, the team can survey roughly 2x as much area as traditional dredge-based techniques in about 1/3 the time while recording information about sea floor topography and water quality, all without damaging marine habitats.

Georeferenced still images are extracted from the video footage, which has resulted in over 250,000 images being recorded over 2011's the 10-day survey campaign. At an AUV speed of 1.5 m/sec, the photos are spaced 37.5cm apart, providing thorough coverage of the sea floor. Throughout the survey runs, the AUV adjusts itself in real-time to follow the contour of the ocean bottom dynamically at an altitude of ~2m (+/- 10cm).

How SubseaObservers Works

After creating a free account and naming their own AUV, contributors are invited to take a brief training mission that provides instructions about how to use SO's interface and how to complete a scallop survey. The scallop tagging tools are very user friendly and provide everything a contributor will require to make accurate observations, including example imagery and a tutorial video. Contributors are also encouraged to file entries in their personal online Logbook, to point out particularly interesting photos and to keep a record of the scallops they've discovered. Contributors must achieve a passing grade in the preliminary training mission before being allowed to survey the dataset to ensure that they understand the task at hand and that they have developed a basic level of proficiency.

As participants progress, they'll climb the Leaderboard to compete with other citizen scientists from around the world (and the project's science team!).

Intended Outcomes

Data submitted to SO will be analyzed and turned into reports that NOAA and other regulatory agencies can use to monitor and set limits. Depending on the speed with which contributors can fully analyze the image dataset, limits could be updated mid-season to allow for more or less harvesting and finer-grained control of stock management. SO will involve the public directly in science activities and conservation efforts, will serve as an engaging medium for communicating findings, and will help to increase interest in STEM while improving scientific literacy.

Project Status

SO was launched in October 2012, and is currently online and ready for use! Sign-up at to get started.

Opportunities for Collaboration

Although SO's current focus is sea scallops, the underlying software has been developed to support many different kinds of marine research. SO is particularly well-suited to any kind of marine field research that involves the collection of datasets that are prohibitively challenging to fully characterize by small teams due to their volume, or by automated systems due to their complexity.

'Mission types' can be developed to support the analysis of photos, video, audio, tabular data, interactive data models and much more. Additional AUVs and other sensor platforms specific to the data being processed can be included in the system so that contributors can build up their own virtual fleet of ocean exploration systems. Moreover, the incremental cost of developing and maintaining a data-specific mission type is significantly less expensive than building and promoting a custom web application from scratch.