Nested Sampling Strategies and a Scientific Information Management Schema for Deepsea Environments

For Presentation at the:
GISPlanet '98
International Conference and Exhibition on Geographic Information

7-11 September 1998

Dawn Wright
Department of Geosciences
Oregon State University

Extended Abstract Submitted: 13 March 1998

Download full paper as an MS Word file
Download full paper as an RTF file

New opportunities continue to arise for geographical information scientists to venture offshore. One of the most pressing is in the field of ocean floor mapping, where we still know less of our ocean floor than we do of the topography of Venus and Mars (99% of which have been mapped in enough detail to reveal features on the order of 50 m in height). In fact, we have made more progress in the past 20 years mapping the surface of neighboring planets than we have in the past 500 years mapping the ocean floor of our own planet . However, the development in the last 10-20 years of sophisticated technologies for ocean data collection and management hold tremendous potential for mapping and interpreting the ocean environment in unprecedented detail. With the understanding that ocean research is often very costly, yet deemed extremely important by large funding agencies, geographical information scientists now have the opportunity to perform coastal and marine studies that are more quantitative in nature, to formulate and test basic hypotheses about the marine environment, and to collaborate with scientists working in corollary subdisciplines (e.g., geomorphologists, political geographers, computer scientists, nautical chart specialists, etc.), as well as with classically-trained oceanographers.

This paper begins by reviewing the newest advances in mapping and management technologies for undersea geographic research (particularly on the ocean floor) and discusses the contributions that geographical information scientists stand to make to a greater understanding of the oceans. During the last 25 years of high-resolution exploration of the ocean floor, there have not only been fundamental, exciting scientific discoveries made, but much has been learned about what combination of tools is needed to fully investigate the interdisciplinary scientific questions at hand. For example, a towed vehicle collecting swath bathymetry and side-looking sonar, towed ~100 m off the bottom at a speed of 1.5 knots, may image ~130 km2 of ocean floor per day. A similar platform with camera/video capabilities, towed ~10 m off the bottom at 1 knot would image less than 1 km2 of ocean floor per day. A manned submersible with a bottom time of 6 hours may traverse 1-3 kilometers of ocean floor during a dive or carry out sampling or experimental tasks at a single location. Given these spatial limitations, a nested investigative strategy is in order: (1) use the high-resolution mapping capability of the towed vehicle to resolve properties of the ocean floor at scale large enough to place the results of a near-bottom investigation into a regional context; (2) deploy a submersible, like a dart aimed at a specific target, to investigate, sample, and characterize a limited number of diagnostic locations with the regional framework defined by the towed vehicle. As effective as this strategy is, it is still limited by the short time scale of the manned submersible investigation.

The last 10-15 years have seen the development of remotely operated vehicles (ROVs) for scientific and industrial applications in shallow coastal waters (less than 1000 m). ROVs have all the characteristics of a towed vehicle with an additional capability to maneuver on a tether for high-resolution investigations and interactive tasks on the bottom. In the last decade ROVs have been aggressively developed for academic applications in deep water. A host of engineering problems unique to working at great depths have largely been solved, and the technological innovations pioneered for shallow-water ROVs (e.g., lights, cameras, thrusters, robotics, etc.) have been adapted by their deep-water counterparts. Like the towed vehicles for deep water, these ROVs are powered by a conducting cable to the surface ship and can carry out scientific missions for several days to weeks on the bottom; like submersibles, these ROVs can remain stationary and carry out complex sampling and imaging tasks. An operator may control the ROV from a distance using either full robotic control, manual control, or a continuous series of combinations between the two. On the whole these ROVs, when well navigated, can carry out mapping tasks at a range of scales that are unprecedented both on the ocean floor and up in the water column. The m- to cm-scale sonar and image-based maps that they can produce are beyond that resolution of the hull-mounted and towed systems and are achieved much more effectively than submersibles. The marine science community is just beginning to ascend the ROV "learning curve" in terms of system configuration and implementation, but clearly the ROV represents a significant step function in humankind's ability to characterize the global abyss.

The introduction of these sophisticated tools has necessitated the development of reliable data management systems for the various data streams. The cost of acquiring the data alone (U.S. seagoing operations run upwards of $25,000 per day) justifies the development of dedicated systems for management and integration of these data. And there is always the goal of using these systems as an analysis tool to optimize scientific interpretations and facilitate the rethinking or reformulating of hypotheses. As mentioned before with nested surveying strategies, bathymetric data from a swath mapping system located underneath a ship may need to be georeferenced to underwater video images or sidescan sonar data collected from a vehicle towed behind the ship and several meters above the ocean floor, to sample sites, observations, temperature measurements, etc. collected from a submersible or ROV launched away from the ship and operating very near the ocean floor, to earthquake data obtained from an ocean bottom seismometer anchored on the ocean floor. The data produced by these different sensors will invariably have different dimensionalities, resolutions and accuracies. And as transmission rates of up to several Gb per day at sea become more and more commonplace, the ability to assess ocean floor data collected at these different scales, in varying formats, and in relation to data from other disciplines has become crucial. Here geographic information science has made a contribution through the introduction of GIS, which fulfills not only the requirement of data integration, but of combining or overlaying data of the same dimensionality. This also serves as an efficient means of assessing the quality of data produced by one instrument as compared to another.

One of the major contributions that GIS stands to make to both global and regional investigations of the ocean environment is, of course, the establishment of long-term mechanisms for automated mapping, data archiving and data distribution. This is especially relevant due to the multidisciplinary nature of oceanographic research. Described in the paper is a scientific information management schema for the integration of accurately-located, high-resolution topographic, photographic, seismic, chemical, geological, biological and acoustic data from the Endeavour Segment (Northeast (NE) Pacific) into a marine GIS for support of a seafloor observatory program. A primary goal of the observatory is to assess the temporal variability of the mid-ocean ridge (or seafloor-spreading center) in the region, within the context of observable tectonic, volcanic, and hydrothermal processes. To achieve this goal, multidisciplinary, coordinated measurements of the ridge crest are continually required at a range of scales. In order to maximize the scientific return from these measurements, steps toward GIS implementation have begun, only the second major effort of this kind for NE Pacific oceanographic data. The goals for the GIS are to provide comprehensive data management, documentation, a web-based, interactive user interface, and data quality control for the multidisciplinary data of the Endeavour Segment. New data conversion schemes and the establishment of the web-based interface are in progress. When complete, the marine GIS will become an essential component of the observatory program by: (1) enabling scientists to integrate several databases into a georeferenced system, where logical queries can be made and spatial relationships assessed between various layers or themes of data; (2) providing a rapid, "unselfish," and logical way to disseminate knowledge, enabling rapid response to such events as seafloor seismic activity, volcanic eruptions and their ensuing megaplumes; and (3) maximizing the scientific return on the millions of Federal dollars spent by the U.S. National Science Foundation, the National Oceanic and Atmospheric Administration, and other agencies involved in collecting data from throughout the NE Pacific.

KEYWORDS: Oceans, Regional to Global Mapping, Ocean Floor Mapping, Bathymetry, Seafloor Spreading, Underwater Vehicles, Information Management and Modelling, Marine GIS, Interdisciplinary Science.

Dawn J. WRIGHT, member of the GISPlanet International Committee

Dawn J. Wright is an assistant professor in the Department of Geosciences at Oregon Statue University, where she is responsible for teaching, research, and campus leadership in GIS. Additionally, she holds an adjunct appointment in the Oregon State College of Oceanic and Atmospheric Sciences and collaborates with faculty and students in the Marine Geology and Marine Resource Management programs. Dawn Wright is a member of GISPlanet International Committee.

Dawn's research focuses on application and analytical issues in GIS for oceanographic, particularly data conversion, management, and metadata. She is also interested in web-based GIS and the geography of Cyberspace. Other research interests include the relationships between volcanic, hydrothermal and tectonic processes at seafloor-spreading centers, and the analysis and interpretation of data from deepsea mapping systems.

Dawn joined the faculty of Oregon State in 1995, having completed a Ph.D. in physical geography and marine geology at the University of California at Santa Barbara and a post-doc at the NOAA Pacific Marine Environmental Laboratory in Newport, Oregon. She has also spent several years at sea as a marine technician for the Ocean Drilling Program and as a shipboard scientist on various research vessels in the U.S. University National Oceanographic Laboratory System.

Oregon State University
Department of Geosciences
104 Wilkinson Hall
Corvallis, OR 97331-5506
Tel: +1.541-737-1229
Fax: +1.541-737-1200

Last updated: September 6, 1998