Cooperative Agreement for CESU-affiliated Partner with Piedmont-South Atlantic Coast Cooperative Ecosystem Studies Unit | G22AS00129

Opportunity Information: Apply for G22AS00129

This opportunity is a US Geological Survey (USGS) discretionary funding announcement for a cooperative agreement under the Cooperative Ecosystem Studies Unit (CESU) Program, specifically tied to the Piedmont-South Atlantic Coast CESU. The project focus is research and development at the intersection of computer science and ecology: building and testing deep convolutional neural network approaches that can recognize individual fish from images. The work is positioned to support and extend ongoing efforts at the USGS Eastern Ecological Science Center (EESC), where researchers are already applying artificial intelligence to identify fish in freshwater streams and to support early detection of diseased fish. The broader intent is to improve ecological monitoring and analysis by pairing modern AI methods with field and laboratory expertise in fisheries science.

The technical goals described in the notice center on three connected deliverables. First, the selected CESU partner is expected to develop deep convolutional neural network methods for image-based individual fish recognition, meaning the models should be able to distinguish one fish from another using visual data rather than simply classifying by species. Second, the work includes developing baseline AI models to establish a performance benchmark for individual recognition, which typically means creating initial reference models and datasets, choosing appropriate architectures and training procedures, and defining standard evaluation protocols. Third, the partner is expected to determine and document performance of the AI models, which implies systematic testing, validation, and reporting of accuracy and limitations under conditions relevant to real-world ecological studies (for example, different lighting, water clarity, fish poses, image quality, and other sources of variation that can affect image recognition outcomes).

Administratively, the award mechanism is a cooperative agreement, which generally indicates substantial involvement by the federal agency compared to a standard grant. The USGS is the awarding agency, and the activity category is Science and Technology and other Research and Development (CFDA 15.808). Eligibility is restricted: applicants must already be participating partners in the Piedmont-South Atlantic Coast CESU network, reflecting the CESU model where universities and other partners collaborate with federal scientists to deliver applied research, technical assistance, and education aligned with agency needs.

Key announcement details include the funding opportunity number G22AS00129, an award ceiling of $99,000, and an original closing date of January 26, 2022. The opportunity was created on December 14, 2021. In practical terms, this is a targeted, CESU-limited research collaboration intended to produce usable AI methods and performance evidence for identifying individual fish from imagery, strengthening ecological monitoring capabilities and supporting related disease detection and fisheries assessment work led by USGS researchers.

• The Geological Survey in the science and technology and other research and development sector is offering a public funding opportunity titled "Cooperative Agreement for CESU-affiliated Partner with Piedmont-South Atlantic Coast Cooperative Ecosystem Studies Unit" and is now available to receive applicants.
• Interested and eligible applicants and submit their applications by referencing the CFDA number(s): 15.808.
• This funding opportunity was created on 2021-12-14.
• Applicants must submit their applications by 2022-01-26. (Agency may still review applications by suitable applicants for the remaining/unused allocated funding in 2026.)
• Each selected applicant is eligible to receive up to $99,000.00 in funding.
• Eligible applicants include: Others.

Apply for G22AS00129

[Watch] Creating a grant proposal using the step-by-step wizard inside the applicant portal: