Water Sampling from Autonomous Underwater Vehicles


DOEI Project Funded: 2005

The advent of in-situ chemical sensors on Autonomous Underwater Vehicles (AUVs) has allowed us to better characterize the structure of chemical plumes. In several cases AUV-borne sensors have revealed fine-scale structures that are not detectable with traditional methods. However, this in-situ data has been viewed with initial skepticism. We hope through this proposal to augment chemical sampling from AUVs with autonomous sample collection and return, to help resolve many of the outstanding questions and skepticism surrounding in-situ chemical sensing techniques.

Hover capable Autonomous Underwater Vehicles such as Seabed and ABE are uniquely suited for conducting in-situ chemical oceanographic measurements.

Specifically they combine

  • An ability to conduct very precise (navigation accuracies close to 1m); repeatable surveys over large areas (10km2) near the air sea interface, in the mid-water column and close to the seafloor.
  • An ability to operate at timescales sufficient to document temporal variability (with mission durations between 10-30 hours) enabling us to study the spatial and temporally varying chemical processes in the neutrally buoyant plumes at hydrothermal vent sites cold seep sites and in costal waters.
  • And finally , due to inherent design characteristics such as the near neutral ballasting of most AUVs, they can move through the water column and approach bottom sediments without creating large turbulent wakes or high-energy thruster wash that would typically prevent us from collecting undisturbed fluid samples

Gas tight water samplers on AUVs should enable the calibration of in-situ instruments that are currently in use (e.g. solid state methane sensors, and optical sensors such as fluorometers and turbidity meters) and provide ground truth for instruments that are under development including the Gemini AUV mounted mass spectrometer.  They should also allow multi-point correlation studies of in-situ data using other types of offsite analyses of the water samples such as through the use of gas chromatographic analysis.  Our design for such a system will allow for triggering based on real-time sensor for determining if local maxima/minima are actually due to heterogeneity or simply sensor artifacts.  Fundamentally such a sampling system will provide a means for more detailed offsite investigations into the makeup of chemically unique areas.

We hope to routinely use this system on deployments at hydrothermal vent sites and cold seep sites and also for sampling the neutrally buoyant plume at the Gakkel Ridge in the Arctic.