Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
2 Lessons from Existing Ocean Observatories Over the past 30 years oceanographers have gained valuable experi- ence with a few pioneering ocean observatories (Appendix D). Some of these have been "observatories of opportunity," in which systems built for other purposes have been leveraged for research. One such program was the network of ship-based Ocean Weather Stations (OWS), estab- lished after World War II for the primary purpose of guiding ocean- crossing aircraft, which ended in 1981. While on station, ships collected oceanographic data that played an important role in early efforts to un- derstand how the ocean has changed over time. In particular, data from this program helped define the relationship between the ocean and the atmosphere. Another example of an "opportunistic" observatory is the Sound Sur- veillance System (SOSUS), a classified system developed by the U.S. Navy in the late 1950s to detect, track, and classify Russian submarines using arrays of underwater hydrophores. SOSUS is a network of acoustic ar- rays in which hydrophores are connected to a shore station by a subma- rine cable. Since the end of the Cold War, oceanographers have been provided limited access to the SOSUS network. Researchers with security clearances have used the system for productive studies of mid-ocean ridge volcanic-hydrothermal systems, marine mammals, and acoustic thermom- etry. SOSUS has also provided the research community with engineering know-how that will be relevant to any cabled network of ocean observa- tories. However, SOSUS has also highlighted the national security con- cerns that will be raised by observatories with similar capabilities. 28
LESSONS FROM EXISTING OCEAN OBSERVATORIES 29 More recently, a small number of long-term measurement sites have been established in both coastal settings and the open ocean using arrays of moored buoys. One of the most successful deep-sea sites has been the TAO array in the equatorial Pacific. Developed as part of the 10-year international Tropical Ocean Global Atmosphere (TOGA) program (1985- 94), the TAO array enabled improved detection, understanding, and pre- diction of E1 Nino events and provided essential data to further under- standing of the E1 Nino Southern Oscillation. Other upper-ocean and air-sea interaction observatory sites that have been established and main- tained in recent years include the Bermuda Atlantic Time-series Station (BATS), the Hawaii Ocean Time-series (HOT) program, and the European Station for Time-series in the Canary Islands (ESTOC) (Appendix D). A small number of cabled observatories have also been established in coastal settings. One of the oldest is the Field Research Facility (FRF) off Duck, North Carolina, established by the U.S. Army Corps of Engineers (USAGE) in 1977. The FRF provided a unique infrastructure of cabled observatory sensors, measurement platforms, and deployment vehicles to support a wide variety of basic research studies of near-shore fluid and sediment processes at an open-coast beach. The 1996 installation of the Long-Term Ecological Underwater Observatory (LEO-15) in 15 m of wa- ter off the New lersey coast pioneered the use of cabled observatory sys- tems for multidisciplinary, integrated studies of meteorological, physical, and biological processes on the continental shelf. In 2000, another cabled, near-shore observatory was established off the south shore of Martha's Vineyard. The Martha's Vineyard Coastal Observatory (MVCO) is being used as a natural laboratory to study how winds, waves, and currents affect the coastline and to monitor oceanic and atmospheric conditions. The Hawaii Undersea Geo-Observatory (HUGO), the first submarine vol- cano observatory, was installed at Loihi Volcano off Hawaii in 1997, using a 47 km electro-optical cable donated by AT&T. The Hawaii-2 Observa- tory (H20), a permanent deep-water geophysical research observatory, was installed in 5000 m of water about halfway between Hawaii and California in September 1998 on the retired Hawaii-2 telecommunications cable (HAW-2~. These observatories have not only demonstrated the significant po- tential of observatory science but also have provided valuable lessons on the installation, management, and operation of ocean observatories. These lessons should be considered as plans are developed for the new genera- tion of ocean observatories envisioned in the OOI. In this chapter some of the 'lessons learned' for each of the implementation issues addressed in Chapter 4 are briefly summarized.
30 ENABLING OCEAN RESEARCH IN THE 2 lST CENTURY PROGRAM MANAGEMENT · Experience from existing observatories (e.g., LEO-15, FRF) indi- cates that observatories must be under the control of their users, the re- search scientists, to provide maximum innovation and flexibility. Experi- ment prioritization should be controlled by scientific requirements and resource availability. · A successful observatory program including well-balanced sci- ence priorities, development goals, and operational needs requires con- stant communication among scientists, engineers, and managers (e.g., TAO, BATS, HOT, FRF). · Existing observatories (e.g., FRF and LEO-15) have proven the value of a broad-based, interdisciplinary approach to observatory research and the need for early and continued involvement of the modeling com- munity to ensure that data collected are of sufficient quality and quantity to be of value to modeling efforts. · Experience from almost every existing ocean observatory has shown the difficulty of sustaining funding for the maintenance and op- eration of long-term observations in a funding environment dominated by short-term, two- to three-year grants. SENSORS 1 tJ · Sensors appropriate for observatory use must maintain their cali- bration and sensitivity characteristics for long periods of time at least six months to a year (so-called 'observatory-capable' sensors). · Years of deploying open ocean (e.g., TAO, BATS, HOT) and coastal moorings (e.g., FRF) have shown that degradation of sensors on surface buoys (due to exposure, vandalism, or sea birds), and damage to sensors in the upper ocean and shallow, coastal waters (due to biofouling, fish bite, corrosion, ship collision, or fishing gear) are major problems that must be addressed to reduce maintenance costs and improve the reliabil- ity of observatory measurements (Figure 2-1~. · Observatories such as LEO-15 and MVCO clearly indicate that ba- sic suites of observations, too expensive for individual investigators to gather, provide the essential scientific context for an observatory's effec- tive use in research and need to be provided as part of the observatory infrastructure. · Results have shown that one of the greatest benefits of ocean obser- vatories (e.g., FRF, HUGO, H20) is high-frequency, long-duration sam- nlin~ that can delineate transient events such as frontal passage, harmful . algal blooms, volcanic eruptions, or severe storms. · Experience demonstrates that the development of new in situ in- strumentation for observatory use is a lengthy process requiring multiple
LESSONS FROM EXISTING OCEAN OBSERVATORIES 31 FIGURE 2-1 This before-and-after picture shows the tremendous problem that biofouling presents to the long-term deployment of instruments in the upper ocean, especially in coastal regions. Figure courtesy of Richard lahnke, Skidaway Institute of Oceanography. cycles of design, field-testing, troubleshooting, and redesign before the instruments become seaworthy enough to be used routinely. · Moorings (e.g., the Bermuda Test Mooring-BTM) or cabled sea- floor junction boxes (e.g., LEO-15 or FRF) should be easily accessible (close to shore) for testing new technology and instrumentation. Such access can greatly accelerate the pace of developing and proving new technology and instrumentation, as well as providing a platform for establishing the comparability of a new technique with older methods that are being phased out.
32 ENABLING OCEAN RESEARCH IN THE 2 lST CENTURY CONSTRUCTION, INSTALLATION, AND TESTING · Experience from the telecommunications industry and at research observatories such as HUGO, LEO-15, and MVCO shows that securing landing rights for a cable shore station, conducting various environmen- tal assessments, and securing the necessary regulatory approvals from local, state, and federal authorities, is extremely time consuming requir- ing lead-times as long as two years. Using retired, in-place telecommuni- cations cables and existing cable stations eliminates the need to go through a lengthy permitting process. · H20 and HUGO demonstrated that reuse of retired telecommuni- cation cables for ocean research observatories is feasible. The principal limitation is power, not bandwidth. · Experience with HUGO and H20 has indicated that engineering instrumentation should be installed on observatory nodes to provide key feedback on infrastructure performance and design and to evaluate in- strument performance. · Experience at HUGO, H20, and FRF emphasizes the importance of knowing seafloor properties and, in some settings of armoring or burying of cables. HUGO failed six months after it was emplaced because of me- chanical wear to unarmored cable lying on the rough volcanic terrain of Loihi submarine volcano. Experiences at FRF demonstrated that strong wave and current action on a cable lying on the seafloor can cause abra- sion of the cable as well as strain on junction boxes, cable splices, and sensor connectors (Figure 2-2~. A cable buried even several meters can become exposed following sediment movement. · Several commercial firms are available for cable installation and ocean hardware design and maintenance. The considerable experience of these firms represents a valuable resource for observatory design and installation (e.g., HUGO, H20) that the academic community should uti- lize. OPERATION AND MAINTENANCE · The OWS, BATS, and HOT time-series sites showed that the value of time-series data depends on the continuity and length of the record. Ocean observatory systems and their instrumentation must be designed for reliable, continuous operation and maintenance costs must be mini- mized if these systems are to be operated for decades or longer. · Observatory operations and maintenance (whether for moorings or seafloor nodes) requires the availability of skilled and experienced personnel. Availability of trained personnel to go to sea is an occasional problem at present and will pose an even greater challenge for the ex- panded observatory network envisioned by the OOI (e.g., FRF, TAO).
LESSONS FROM EXISTING OCEAN OBSERVATORIES 33 FIGURE 2-2 Seafloor junction box at HUGO, atop Loihi Volcano off Hawaii, after five years on the ocean floor. The partially buried package, recovered in October of 2002, is still in excellent condition. Use of titanium and plastics has virtually eliminated corrosion problems. Figure courtesy of Fred Duennebier, University of Hawaii. · Coastal observatories are most beneficial when an inventory of community coastal vehicles (e.g., skiffs, surf zone working platforms, etc.), are available on site for deployment and retrieval of sensors (e.g., FRF) (Figure 2-3~. · Surface observatory nodes require regular routine servicing to re- place or repair system components and instrumentation and to maintain data quality and continuity. Experience shows that these maintenance costs are often initially significantly underestimated (e.g., LEO-15, FRF). · Even with the few open-ocean observatories currently in opera- tion, the limited availability of ships and ROVs has hampered their opera- tion and maintenance (HUGO and Hem. These assets are in high de- mand and scheduled well in advance, making it extremely difficult to respond quickly to observatory failures. The future need for ROV assets
34 ENABLING OCEAN RESEARCH IN THE 2 lST CENTURY FIGURE 2-3 The USACE FRF's Coastal Research Amphibious Buggy (CRAB). This deployment and retrieval platform is a tower on three wheels, hydraulically driven by a diesel engine located on the operations platform, 11 m above the seabed. The CRAB can operate in 2 m high breaking waves in 10 m depth of water, with Global Positioning System (GPS)-positioning accuracy to deploy flu- id- and sediment-boundary layer sensors. The CRAB has also been used to mea- sure monthly bathymetric changes at the Duck, NC site over the past 25 years. Figure courtesy of loan Oltman-Shay, Northwest Research Associates, Inc.
LESSONS FROM EXISTING OCEAN OBSERVATORIES 35 will likely increase significantly as more ocean observatories are estab- lished. · Experience from years of operating oceanographic moorings shows that inter-comparison and inter-calibration procedures are essential. These procedures ensure data are of uniform quality even though they are col- lected from diverse sites using different instruments. Ship and/or ROV time must be dedicated to in situ sensor performance checks just before and just after deployment, as well as on recovery; a requirement which is often overlooked. NATIONAL SECURITY · One lesson from the U.S. Navy's SOSUS array is that the acquisi- tion and public distribution of acoustic and other geophysical data in some regions along the U.S. coastline poses a significant national security risk. Deploying sensitive arrays in some areas could lead to the need to restrict data access, prevent data acquisition at random intervals, or re- strict publication of results. DATA MANAGEMENT · Other major ocean science programs such as RIDGE, the World Ocean Circulation Experiment (WOCE ), and the Joint Global Ocean Flux Study (IGOFS) show that the ocean observatory program cannot rely on individual investigators to manage, archive, or disseminate observatory data. Instead, data must be professionally managed and distributed through established data centers according to a policy that guarantees data is made available to the science community. · Experience has demonstrated the value of establishing data for- mats and metadata content among participating data centers during the design phase of an observatory before data collection begins. For example, H20 data is distributed in the Standard for the Exchange of Earthquake Data (SEED) format, making it particularly valuable to seismologists around the world. · Major ocean science programs like RIDGE, WOCE, and JGOFS in- dicate that strong interaction between science teams and data manage- ment groups is essential for a data management system. Without this interaction, requirements are often not implemented as intended. · Instrument interfaces and data formats should be compatible amongst various instruments to simplify their integration. A standard- ized interface that would automatically associate an instrument's meta- data with its data stream would facilitate such integration.
36 ENABLING OCEAN RESEARCH IN THE 2 lST CENTURY · Considering the large data acquisition rates (e.g., Gb/s to Tb/s), and the complexity of data products that an ocean observatory network could produce, levels of data products need to be defined and standard- ized for sharing data. For instance, both satellite missions and Argo define multiple levels of data products. · Data archive centers require sustained funding to support data archival and distribution, even after the life of a program. For example, the Incorporated Research Institutes for Seismology's (IRIS) Data Man- agement System archives approximately 3.5 Tb per year of seismic wave- form data and is funded at about $3.5 million annually. In addition, NASA maintains three major archive centers and one long-term backup center at a cost of close to $100 million a year. EDUCATION AND PUBLIC OUTREACH · Experience has shown that education and outreach efforts will be more successful and more cost-effective if they are part of the initial ob- servatory design rather than an afterthought. · As illustrated by experience with H20, outreach is best if handled at a central facility (in this case, IRIS) than by individual investigators. · NASA has shown that a successful education and outreach pro- gram requires a professional staff with expertise in education and out- reach. These activities need to be coordinated at the program level, not at the individual investigator level. In addition, a mandatory percentage of every project budget should be earmarked to support education and out- reach activities.
