Applications of Analytical Chemistry to Oceanic Carbon Cycle Studies (1993)

Ocean sensors have the potential to measure a variety of analytes dissolved in seawater, particularly with the combination of a variety of transduction mechanisms. The ultimate instrument package could combine sensors based on electrochemical, mass, optical, and piezoelectric mechanisms, all employed for measuring different parameters. No single technology will be useful for solving all measurement problems. Nonsensory types of instrument packages, such as gas chromatographs, capillary zone electrophoresis, and mass spectrometers, will continue to be of some value in addressing ocean measurement needs in the future. Extensive in situ measurements are vital for calibration of measurements made remotely from aircraft and satellites. It is impractical to make frequent and detailed measurements from ships, due to lack of both physical and human resources to undertake such an immense task. Sensors will provide the capability to discover many previously unobserved chemical processes and to achieve a better understanding of the ocean system.

As discussed in the introduction, there are compelling reasons to study the ocean. The role that the ocean plays in the world's climate, pollution, and food resources can be understood fully only with appropriate measurement capabilities. The development of chemical measurement technologies is critical to understanding both the large-and small-scale processes of the ocean. The U.S. government must take the lead in supporting the research

and development effort for ocean instrumentation because the government, through its agencies, is the beneficiary and ultimate consumer of most ocean measurements. Consequently, it must facilitate the advances and implementation of new measurement methods, through funding agency, commercial, and academic research and development. The committee concluded that the market for ocean instrumentation is too small to stimulate significant commercial interest, so that government support (both financial and administrative) is also required to encourage companies to produce new instruments. The uniformity, documentation, and engineering support afforded by commercial development is critical to widespread acceptance of new technology. One funding vehicle that already exists and has been used to some extent is the Small Business Innovation Research (SBIR) programs. Agencies that sponsor or carry out ocean science measurements should include instruments as examples in their list of desired research for SBIR programs. SBIR programs are too limited in time and total funding, however, to sustain the development of some instruments.

An instrument development program has a 10-year development cycle, as discussed earlier in this report. The nature of ocean measurement instrumentation requires interdisciplinary programs that tend to require longer-term commitments than do single-discipline efforts. These programs require a long time to test and prove the suitability of the instrumentation to the particular measurement need. Many iterations may be required, including extensive ship time to allow researchers to develop confidence in a new measurement technology. The ability to communicate remote data via satellite and other means must also be improved for maximum advantage from chemical, as well as other types, of sensors. Acoustic telemetry adds some promise for underwater data transmission, although it is limited by the variability of the acoustic medium at the long wavelengths that must be used.

Ocean measurements have global impact. Consequently, federal involvement must occur at an international level as well. The International Oceanographic Commission is promoting the idea of a global ocean observing system, which would eventually require new chemical and biological sensors. Various agencies must be involved in coordinating this international effort, including the Department of State, as ocean measurement needs are international in scope. Finally, the Fulbright Program, as well as NATO fellowships, should be used to disseminate new measurement technologies and research expertise throughout the world community.

Regardless of the approaches taken, it is clear that a considerable investment of time and resources will be required to develop the arsenal of chemically selective and stable host compounds that will be required to

measure all the key oceanic species via in situ chemical sensor technology. The committee recommends that resources be made available to attract specialists in synthetic organic chemistry and polymer materials to the field of selective recognition chemistry for ocean measurements. For telemetered data, processing close to the sensor should be emphasized, to reduce the size of the data stream.

The ability to make analytical measurements depends intimately on the availability of well-defined standards and calibrants. Many measurements of analytes in seawater (such as DOC and DON) cannot be compared among laboratories because of the lack of appropriate reference materials and blanks for instrument calibration and testing. Intercomparison exercises are critical (Williams, 1992). Seawater is a highly complex medium that makes development of standards particularly difficult. The committee recommends that agencies supporting new ocean measurement technology, such as the Department of Energy (DOE), the National Science Foundation (NSF), the National Oceanic and Atmospheric Administration (NOAA), the National Institute of Standards and Technology, and the Office of Naval Research (ONR), devote a portion of their budgets to support development and maintenance of standards and calibrants. Without standards it will be impossible to develop, let alone deploy, new measurement techniques. Quality control for oceanographic data is crucial as the number of laboratories scattered around the world, both shipboard and land based, measure analytes in seawater. It is unfortunate that funding for production of standards and for calibration is often the first to suffer when agency budgets are cut.

Perhaps the major thrust of this report is that improved ocean instrumentation will be necessary to acquire the data necessary to evaluate global change. As noted by Wunsch (1989), the present process and resources for developing ocean instrumentation are inadequate. Present funding for ocean research is ill suited for instrument development because the funding process provides support for a period of 3 to 5 years. Yet ocean instrumentation takes longer to develop than typical laboratory instrumentation, because it must be made to work reliably in a hostile ocean environment after being demonstrated on the laboratory bench. For instance, constructing a specialized instrument for use on land might take 2 to 4 years before publishable results are obtained. For an in situ device, an additional 3 to 5 years might be required. Thus, few results will be obtained before funding lapses. Moreover, there is overt technical risk in developing new instru-

mentation, where both sponsor and scientist may have nothing to show for the effort. Contrast this with studies employing existing methods that provide results immediately and continuously for both sponsor and scientist. Clearly, the time scale and goals of existing programs are antithetical to instrument development. To achieve efficiency in instrument development, sustained involvement of instrument engineers should be pursued. The most common approach now is to employ engineers for the duration of a project. A longer-term approach, as used in the past by NOAA and NSF, should be renewed.

The emphasis of most sponsored research is for new knowledge, which limits the availability of ship resources for instrument development. The importance of ruggedness and reliability for in situ, moored, and autonomous instruments translates into a great need for repeated testing in their development. Ocean instruments are analogous to spacecraft in their need for reliability in a hostile environment. For instrument development, testing is an iterative process, with many trials of different components, configurations, and designs typically performed before adequate performance is achieved. Thus, testing ideally is carried out frequently; under reproducible, well-understood conditions; and often on short notice. For ocean instrumentation, the most important part of development is testing the device at sea. The rate limiting step is availability of ship time. Whereas most instrument development has been carried out on large ships, this is not the most efficient means of instrument development. Unfortunately, ship time (precious at $20,000 a day for transoceanic vessels) is mainly allocated for the acquisition of new knowledge during extended cruises. Such extended cruises are far from ideal for testing, since conditions vary with the different places ships go; the extended cruise does not permit quick return to the laboratory for modifications; the operational aspects of the cruise (course, speed, use of other equipment such as depth sounders) may be ill suited to testing a particular device; and the expeditionary nature of cruises requires months of preparation. Oceanographic cruises are often planned at least a year in advance. Despite these drawbacks, much instrument testing is done on cruises simply because no other option is available.

The problem of access to ship time is more difficult for commercial enterprises than for academic ocean scientists, the latter having some fraction of institutional ship time earmarked for their use. For specialized ocean instrumentation, which usually has a small market, the cost of development is prohibitive if ship time is paid for directly. As is the situation for development of instruments by academics, a large portion of the ship time for commercial instrument development is obtained on craft of opportunity or piggybacked with academic research. The difficulty of doing this varies with the sponsoring institution. A few larger companies able to operate their own ships can spread operating costs of the ships over several devel-

opment programs. It is crucial to note that a large fraction of the high cost of development is attributable to the duration of the development process. For instance, a 5-year development program accomplished in 3 years would save 2 years of salary and benefits of the participants. Clearly, improving access to ship time for instrument developers would enhance availability of ocean instrumentation.

There are a range of approaches which might be taken to expedite instrument development. One would be to establish research vessels or moored buoys whose primary mission would be to serve as test platforms for instrumentation. Instead of embarking on extended cruises to scattered destinations, these ships would operate in a few areas for which oceanographic data are available in abundance. These ships would be operated on a short-notice basis, with maximum flexibility for handling different sorts of equipment. Repeated visits to the same areas, probably in near coastal areas, would enable moored devices or buoys to be repaired, recalibrated, or returned to shore easily. These ships could be made available for use by academics, government personnel, or commercial ventures in instrument development; this broad use would leverage the resource greatly. Ship scheduling would be controlled by some central authority, such as the University-National Oceanographic Laboratory System (UNOLS) that is presently used to schedule regular oceanographic cruises. If instrument development were viewed as a national priority such that sufficient resources were made available, specifying ships for this purpose should be considered.

The use of buoys for instrument testing also shows great promise. Buoys that are part of the Joint Global Ocean Flux Study are located in the Atlantic and Pacific oceans for the purpose of regular measurements over long time periods. These buoys are visited by ships on a regular basis and are being used for long-term testing of instruments.

Even without additional resources, much can be done to expedite instrument development. Ship time requests for the purpose of instrument development might be coordinated by UNOLS to make more efficient use of the available vessels. Government-funded vessels might be utilized by commercial ventures as well as by academics, if some equitable prioritization and cost sharing system were developed. Other means for expediting testing might be sought, including making more vessels of opportunity available when these are suitable. Navy and Coast Guard ships might be used when this does not conflict with their primary missions. Cooperation for instrument testing already exists informally; what is required is a mechanism whereby the respective bureaucracies can be encouraged to cooperate. Similarly, the realization that development takes longer and has goals and milestones different from those of ocean science should encourage sponsors of ocean science to restructure their support. In particular, the funding periods must be extended and greater provision made for testing.

The research enterprise is better served by identical, well-documented devices throughout the world than by sporadic availability of unique, poorly documented devices; consider the impact on optics research if scientists were obliged to build their own lasers, or on physical oceanography if each scientist built profilers for conductivity, temperature, and depth rather than buy standard models from commercial vendors. While it will still be justifiable to develop instrumentation for a single use, those seeking support for developing widely useful instrumentation should demonstrate that developed instruments have the potential to move beyond proof of principle.

The committee chose to categorize a number of analytes in terms of their importance for measurement in the ocean in the coming decade. The selections reflect the needs of the major oceanographic research programs, particularly for understanding the ocean's role in the global carbon cycle, and include some other analytes associated with other issues of great scientific interest. In order to make an interdisciplinary effort between oceanographers and analytical chemists work, oceanographers must be willing to identify a limited number of analytes on which analytical chemists can begin to focus. This report focused on understanding the global carbon cycle, but the committee does not suggest exclusion of other areas from research support.

The committee recognizes that successful implementation of an ocean instrumentation development effort requires extensive ongoing transfer of information and opportunities between the ocean science and analytical chemistry communities. The committee recommends a multipronged approach to accomplish this information transfer. In addition, the committee believes that an effort should be undertaken to educate both the public and promising students about the opportunities in this exciting field. Potential vehicles for accomplishing these goals include the following.

1. Organize miniconferences (for example, a ChemRawn conference; e.g., Goldberg, 1988) or 1/2-to 1-day symposia at major analytical chemistry and ocean science conferences (for example, the Pittsburgh Conference, Gordon Conference, and American Geophysical Union) devoted exclusively to discussing the recent advances and challenges in developing ocean measurement technologies.

2. Explore the establishment of a true organizational ''home'' for analytical oceanography, preferably as a new subdivision within an already existing American Chemical Society Division (for example, Analytical Division, Environmental Chemistry, and Geochemistry).

3. Important measurement issues must be communicated from the ocean science community and new analytical advances to the ocean science community. The effort should involve articles in professional and trade journals and conference presentations for both communities. These mechanisms could improve transfer of information between ocean scientists and analytical chemists.

1. The need and opportunities for analytical research relating to ocean measurements could be publicized through articles in key journals or magazines (Analytical Chemistry: e.g., Johnson et al., 1992, C & E News, Journal of Chemical Education, and others).

2. Information about new measurement technologies could be publicized to oceanographers through EOS, Oceanography, Sea Technology, and other publications.

3. Videos and brochures describing the kinds of measurement tools and techniques required to study and understand ocean chemistry could be prepared by agencies and societies that would benefit from the development of new ocean measurement technologies.

4. Undergraduate chemistry curricula can be developed that expose students to concepts of environmental chemistry as they relate to current global oceanography issues (for example, global warming and coastal pollution).

4. Interactions between practitioners of ocean science and analytical chemistry can be encouraged by establishing new funding initiatives at government agencies (for example, NSF, ONR, NOAA, and DOE) to foster interdisciplinary research between oceanographers and analytical chemists on timely ocean measurement problems. The academic reward structure for analytical chemists depends on contributions to chemistry, not marine science. Likewise, marine scientists are rewarded for contributing to marine sciences, not analytical chemistry. The availability of appropriate reward structures will affect whether or not the hoped-for results of publicity, education, and training are achieved. These might include

1. New research initiatives in auxiliar areas of fundamental importance to the future of ocean measurement instrumentation (for example, anticorrosion, antifoulants, and recognition chemistry).

2. Opportunities for academic analytical chemists to attend summer workshops at oceanographic institutions for the purpose of learning more about today's ocean measurement technologies. Based

on this first-hand exposure, these scientists could incorporate more examples of oceanographic measurement techniques into their undergraduate and graduate analytical curricula. In addition, such an experience might stimulate more research initiatives by these faculty in ocean measurement technologies.