capabilities oftentimes had to be accomplished over a succession of different proposals, reviews, and awards under sponsorship of different agencies and programs.

In 1981, the Office of Technology Assessment (OTA) reported that technology development across the federal ocean programs was poorly coordinated and was provided mainly through specific objectives of mission-oriented agencies such as the Navy, the National Oceanic and Atmospheric Administration (NOAA), and NASA (OTA, 1981). NSF was shown to have a minimal role in ocean instrumentation and technology development. Research programs were attributed to whatever technology support was provided on an ad hoc basis.

About this same time, observers of the NSF ocean science peer-review process noted that in matching available resources to highly rated proposal budgets, instrumentation development was one of the first items to be eliminated. The focus was on research, more than on new tools to accomplish it. This was especially true of multidisciplinary instrumentation. Funding pressures and the conservative nature of the peer-review process required that NSF-sponsored technology development for basic ocean research either be essential for the accomplishment of the highest-rated research projects or be done at no cost to NSF.

Given these somewhat subjective observations, an experimental program area was established in fiscal year 1982 to consider proposals for developing new instrumentation and new technological capabilities that would have broad applicability. The Oceanographic Technology (OT) program was established within OCFS as part of an overall reorganization of OCE. The OT Program also assumed responsibility for supporting shipboard technicians, the acquisition of commercially available shared-use research instrumentation, and the development of new instrumentation and technology by individual investigators. In keeping with the multiuser facility responsibilities of OCFS, initial proposal submission guidelines for technology development emphasized data collection and general-use instrumentation.

Since this was a new program area and the first of its type for ocean science at any agency, there was a lot of latitude in the scope of the original proposals. Ocean science instrumentation development proposals had to satisfy two major proposal requirements: technological or engineering quality and ocean science relevance. Bimodal ratings occasionally resulted when scientists were enthusiastic about a proposed new measurement capability, but engineering reviewers judged that the proposal was technically flawed. The opposite also occurred when a proposed new development was well reviewed from the technical side, but the science reviewers found the scientific relevance or utility of the new device to be lacking.

From its inception in fiscal year 1982 through fiscal year 1998, slightly more than $55.5 million has been awarded for supporting more than 150 ocean science instrument development projects. Three general categories of projects have been supported, reflecting different community requirements: (1) demonstration projects that typically seek part-time support for a technician or engineer, plus supplies to test an idea for enhancing existing instrumentation; (2) implementation projects that span a range of activities for further developing or modifying existing instrumentation for general ocean science research applications; and (3) instrumentation systems development, which involve major projects, represented by cooperative efforts between scientists and engineers to integrate several instruments and technologies into an observational system. Parallel advances in theory and instrumentation are usually necessitated. Bioacoustic and satellite remote sensing, long-term moorings, tomography, autonomous underwater vehicles, conditional sampling devices built around knowledge-based systems, and fiber-optic sensors are examples of this complex category of development project. A long-term effort is required at relatively high annual cost, and risk of failure is a further consideration.

The peer-review system does not lend itself well to long-term, forward-looking projects with a significant risk of failure. However, to develop new capabilities that are driven by scientific needs, risk can be reviewed and managed. A case in point is the development and establishment of long-term seafloor observatories. The scientific need to make long-term measurements, in both the coastal zone and the deep sea, coupled with newly developed sensors and other technologies, has set the stage for a new way of conducting certain types of ocean science research. The nature of these observatories suggests that they will have to be a new type of facility. However, as with other facilities, their long-term support and viability will depend on their ability to provide the technological capabilities that will be needed to support ongoing ocean science research.

If one considers the phenomenal advances that the academic ocean science research community has made in the past several decades, sponsored primarily by NSF, one may conclude that the provision of technology and the development of new capabilities have been appropriately addressed. An adequate mix of ships and facilities has been provided to the community, research projects have been underpinned by a growing technological base, and OCE has provided funds to lay the groundwork and develop new capabilities for research envisioned in the future. OCE has effectively met a community requirement for supporting projects to enhance and upgrade existing observational and analytical research capabilities. The availability of significant levels of funding for long-term development of new instrumentation and technology should remain a priority for growth. Just as past progress has been based on collaborations with other agencies and endeavors, establishing partnerships and maintaining them are going to be critical for future progress.