Boundary Layer Dynamics (1997)

The panel believes strongly that a number of important synergistic opportunities could enhance boundary layer dynamics research and offers several suggestions toward developing this potential. NRL has the potential to continue as a major contributor to both theoretical and observational boundary layer and remote sensing research in the United States. It has a wealth of facilities and human talent; many of its resources are the envy of university departments. To achieve this potential, however, NRL must effectively integrate and coordinate these resources.

The present status of NRL work related to marine boundary layer dynamics has been conditioned by the historical fact that naval research in oceanography, from the mid 1970s on, was to be performed by the former Naval Ocean Research and Development Activity (NORDA) in Bay St. Louis, Mississippi, while new developments in naval meteorology were to be pursued by the former Naval Environmental Prediction Research Facility (NEPRF) in Monterey, California. Although NRL-E had some ongoing activity in each of these areas, the principal work carried out at NRL-E that related to the marine boundary layers was in remote sensing, both electromagnetic and acoustic. Although these three institutions have been combined administratively into the single Naval Research Laboratory, the physical separation of the NRL's work in oceanography, meteorology, and remote sensing remains largely intact. The challenge is to focus this work intellectually.

The air-sea interface and its adjoining boundary layers represent a common meeting point, both physically and intellectually, for the work of the scientists located at the three sites. However the phenomenological scales, spatial and temporal, of interest at the three NRL sites differ significantly (Figure 3.1). The success of interunit collaborations will depend strongly on how well cooperating scientists from all the sites are able to define and focus their efforts on multiscale and intersecting-scale joint research projects. Formation of a new center as an organizational focus to help accomplish this integrated effort is addressed in the next section.

For successful collaboration, it is imperative that effective and efficient communication be established among the different groups at NRL's three sites. Although communication is made difficult by the geographical separation of the respective units, use of electronic mail and networking can assist collaboration. Regular laboratory-focused scientific meetings should be encouraged. One of the main objectives of these meetings should be to identify a few specific areas for focused research and then concentrate on these in a coordinated way. These areas should involve a mix, appropriate to NRL, of mission-oriented and independent, unsolicited research. This suggestion is discussed more fully below.

Once such NRL focus would involve collaboration in the instrumentation area. This could have a major impact because, ultimately, advances in BLD are very likely to depend on innovations in instrumentation, leading to naval field measurements.

Another specific step toward coordination of efforts at NRL would be to incorporate, in models of both the atmosphere and ocean, the presence of ocean surface waves, as mentioned in Chapter 2. This step could lead to improvement of the models as well as bring recognition to NRL-E, -W, and -S. The necessary data sets exist, from either satellite remote sensing or wave model predictions, and

Figure 3.1 The phenomenological scales currently addressed in boundary layer dynamics research at NRL's three sites. A new center would serve to integrate the combined assets.

it would not be difficult to begin testing the impact of the sea state on either model. For example, the literature describes numerous approaches to including sea-state effects in specifying the aerodynamic drag coefficient, such as parametric MABL models, which require integral parameters of the sea state, and spectral MABL models, which require detailed descriptions of the directional wave field. Most of these models are constructed from limited data sets derived from buoy measurements. None of them have been rigorously tested in a global and/or regional model to determine the effects of and sensitivities to sea-state inputs. Similarly, the ocean modelers could test the sensitivity of their models with improved predictions of the wind stress.

As a result of its consideration of capabilities at the NRL complex, the panel recognized that an unusual opportunity exists to focus the advanced research in BLD at NRL-E and NRL-S to enhance the Navy forecasts produced at the Fleet Numerical Meteorological and Oceanographic Center (FNMOC)/NRL-W facility.

Historically, NRL-W's role has been to support the efforts of FNMOC. Its work on the FNMOC atmospheric forecast model has been of major importance. Another applied activity involves making the forecasts available and useful to the fleet. Thus, NRL-W, as inherited by NRL, seems at present to be primarily a service facility. Service should continue to be one of its primary responsibilities; however, with one of the largest collections of marine meteorologists in the country, NRL-W has the potential to be a valuable national resource.

In fact, NRL-W's access to the huge FNMOC database represents an enormous and unique advantage, and present trends in declassification of databases should facilitate broader cooperation, generally, in its use for research. This capability, coupled with the advantages that come from having a group of modelers that must work in continual contact with an operational user organization, makes NRL-W a special facility. These advantages should not be lost.

Further consideration of NRL capabilities for BLD research led the panel to conclude that there is an outstanding opportunity to focus this research by formation of a new center, located at NRL/FNMOC and modeled after the European Centre for Medium-range Weather Forecasting (ECMWF), where NRL personnel could work, together with researchers and investigators from academia, on global and high-resolution regional coupled models of the atmosphere and ocean. Such a center would have several purposes: (1) to ensure that the operational side of the Navy would be aware of, and understand, the newest developments in the models, and provide for transfers of specific technologies; (2) to give NRL personnel the opportunity to take the lead in certain areas or at a minimum, remain up to date on current problems being addressed in academic research; and (3) to increase academic researchers' awareness of the complexities of naval operations.

ECMWF is one of the world's model facilities that strongly fosters the interrelationship between operational activities and research performed by member research and academic institutions. No similar facility currently exists in the United States. However, NRL-W with its ties to FNMOC and several surrounding academic institutions, e.g., Stanford University, the University of California at Berkeley, the Naval Postgraduate School, and the University of California at Santa Cruz, would be in a unique position to establish a center similar to the ECMWF facility. There are several reasons for doing so.

First, the atmosphere and ocean can no longer be treated as separate, disconnected entities. Much research has been devoted to establishing linkages and interactions among these environmental components in order to improve capabilities for predicting numerous processes such as marine and severe weather, ice movement along the marginal ice edge, and the evolution of the mixed-layer and surface waves on a global scale affecting the Navy's daily operations across a broad spectrum. The suggested center would also provide a means to explore the feasibility of assimilating a variety of remotely sensed data sets and perform end-to-end tests of a combined model of the atmosphere-ocean system.

Second, to make progress in a timely way, operational personnel and researchers from academia and government laboratories must work side by side (at the same facility) in a collaborative mode to exchange ideas and thoughts that then would be evaluated in research and prototype models of operational forecast and prediction models. These research and prototype models would run in parallel, or reprocess historical data sets, to assess the impact of new coupling and feedback schemes and/or assimilation of heterogeneous data sets. At regular intervals, operational models could then be updated with those algorithms and components that are robust and computationally efficient, and could be used to generate either an improved product or new output for use in other relevant models.

Third, the new NRL-W facility could be co-located at an operational facility, FNMOC, where a wide diversity of expertise exists from government and academic institutions (i.e., those in the greater San Francisco area). An active visiting scientist program could complement and/or enhance local expertise. Such a center would have to maintain very high standards and provide leadership in many research areas in order to attract leading scientists and researchers from government and academic institutions. Young and promising scientists would benefit from closer interactions with leaders already working at the center, whereas established scientists would appreciate the problems and complexities associated with operational modeling and prediction. Thus the new center would provide an environment encouraging direct and rapid transfer of concepts and models between academia and government.

Fourth, the center would have to provide adequate access to high-speed computers and resources such as powerful workstations, peripherals, networking capabilities, graphics, and video software. Also, the center should

maintain an adequate permanent staff of researchers, programmers, data processors, and systems analysts for support to visiting scientists. The duration of visits should be no less than six months and not longer than two years. All research work should be unclassified but should not compromise FNMOC operational and/or classified activities. For example, operational and classified activities should involve separate computers and strictly partitioned networks, and access to operational and classified locations should be restricted. ECMWF, the model for the new center, in fact operates in this way. However, the products and methods of the operational models should be unclassified within the new center, to encourage discussion and different approaches to interpretation of the results. There must be several common areas where visiting personnel can interact, exchange ideas, and discuss problems, as well as mechanisms (e.g., seminars, discussion, and small task groups) to explore existing problems with operational models and output products and to evaluate new model components and reprocessed historical data to maintain leading-edge modeling capabilities.

As a result of the general reduction of federal budgets, and of NASA's and NOAA's in particular, NRL's programs will have greater national importance and a national "critical mass" of science in the areas discussed above may be difficult to achieve. The possibility of joint interagency activity directed to the advancement and use of remote sensing for boundary layer dynamics research should be seriously considered.

One way to encourage atmospheric modelers, air-sea investigators, and others to couple their work more directly with remote sensing needs and capabilities of the kind required by the Navy would be to have periodic (annual or semiannual) internal NRL meetings at which the various research groups would present their current work, plans, and problems to one another. These meetings could be run at the scientific level expected at an open scientific meeting but be internal to NRL. Research in progress could be discussed candidly, specific naval priorities could be considered, and plans could be made to direct future efforts toward greater interaction and coordination in this research. In addition, talks by outside experts, and workshops including outside participants, could be encouraged.

NRL currently has in place programs to involve outside scientists in NRL activities through their appointment as summer faculty, consultants, and temporary personnel. It also has a variety of postdoctoral programs. The panel strongly encourages active recruitment of postdoctoral scientists, development of increased contact and collaboration with outside scientists, and an emphasis on visiting appointments for scientists working in BLD research areas.

A possible difficulty in straightforwardly implementing this suggestion is that, in oceanography at the present time, postdoctoral positions are relatively abundant, and NRL is unlikely to have much success in recruiting at this level. However, there are insufficient positions at the end of the postdoctoral time of typically two years. A possible strategy would be to identify and contact oceanographers as they finish their postdoctoral work and to offer them longer-term "visiting scientist" positions with term commitments on both sides. Some of these scientists could then become permanent staff, and the others could continue their collaborations after moving on to other positions.