Award: OCE-2123560

The mesopelagic, or midwater, region of the ocean is a vast and complex environment where physical processes, environmental gradients, particles, and biological communities interact across a wide range of spatial and temporal scales. Despite its importance to marine ecosystems and the global carbon cycle, this region remains difficult to observe. Traditional ship-based sampling methods, including water-column profiles, nets, acoustic surveys, floats, gliders, and towed instruments, measure different components of the ecosystem at different places or times, and synthesizing their findings has been difficult. The goal of this project was to address this challenge by developing and testing a novel midwater profiling vehicle capable of collecting physical, visual, and acoustic observations simultaneously while complementing existing ship-based oceanographic operations.

The VAMPIRE system was designed to repeatedly profile the water column while carrying a complementary suite of sensors. These included multi-frequency acoustics, optical systems for observing organisms and particles, and environmental sensors measuring properties such as temperature, salinity, oxygen, chlorophyll, and optical backscatter. Bringing these measurements together provides an opportunity to characterize not only where organisms and particles occur, but also the environmental conditions surrounding them.

A major outcome of the project was the development, testing, and validation of approaches needed to interpret these integrated observations. The optical component, a high-frequency Underwater Vision Profiler (UVP6), was deployed extensively across a variety of oceanographic conditions, generating observations of particles and zooplankton across depths, seasons, day-night cycles, and geographic regions. These deployments established a substantial reference dataset while allowing researchers to develop and refine workflows for instrument deployment, image processing, annotation, and data analysis and to compare optical observations with conventional net-based measurements of zooplankton.

The integrated VAMPIRE system was deployed alongside multi-frequency acoustic observations and MOCNESS net sampling. This combination produced complementary optical, acoustic, environmental, and organismal datasets that can be compared with established methods for determining zooplankton abundance, size, biomass, taxonomy, and vertical distribution. These sampling plans, which explored a range of deployment patterns including profiling and stationery as well as vertical and horizontal sampling prioritization, provide an important foundation for evaluating how autonomous and imaging-based observations can complement traditional net sampling and for integrating multiple types of measurements into a more complete representation of midwater ecosystems.

The instrument and sampling efforts performed during the grant address the project's intellectual merit by advancing an approach to a fundamental challenge in biological oceanography: simultaneously enumerating, identifying and determining the size of organisms and particles together with the environmental conditions that structure their distributions. The resulting capabilities and datasets can support questions concerning the spatial and temporal variability of organisms and particles, diel vertical migration, relationships between particle distributions and zooplankton patchiness, and the influence of environmental structure on animal distributions. Repeated UVP observations also produced a substantial dataset describing vertical and temporal variability in particles in the Sargasso Sea. Analysis of these observations has characterized different particle types and their relationships with biological and environmental conditions and has contributed to scientific presentations and graduate research.

The project also produced important broader impacts through education, workforce development, and interdisciplinary training. Graduate students gained experience spanning ocean engineering and biological oceanography, including oceanographic instrumentation, research cruises, net sampling, image and acoustic data analysis, interdisciplinary collaboration, scientific communication, and mentoring. A high-school intern received intensive training in oceanographic image processing, data quality control, annotation, and visualization and subsequently became a co-author on scientific presentations.

Project expertise was also incorporated into hands-on engineering education through collaboration between the Bermuda Institute of Ocean Sciences and the University of Rhode Island. Through the Mid-Atlantic Robotics IN Education (MARINE) program, over the course of five summers high-school students participated in activities emphasizing soldering, coding, waterproofing, computer-aided design, underwater robotics, and prototype development. Students applied these skills to authentic ocean-observing challenges and reported increased confidence in engineering skills and greater understanding of how robotics and engineering can address environmental questions. Educational modules developed through the project continued to be used beyond the period of direct project support, extending the impact of the award.

Overall, the project advanced an integrated approach for observing the poorly sampled mesopelagic ocean while producing valuable datasets, analytical workflows, interdisciplinary expertise, and educational resources. By bringing physical, optical, acoustic, and traditional biological observations together, VAMPIRE provides a foundation for more comprehensive investigation of midwater ecosystems and for future development of autonomous approaches to understanding this important region of the ocean.

 

Last Modified: 09/17/2026
Modified by: Amy E Maas