National Aeronautics and Space Administration

Wallops Flight Facility

Ocean Observation


Our ocean coastal regions are an enormous national resource, providing our nation with a source of food, recreation, transportation, waste disposal and acting as a natural boundary. Because humans reside primarily along the coasts and rely on it for food and other resources, protecting it from human influence, climate changes, and ecosystem perturbations is a priority item. The primary focus of the OOSA group is to develop a Regional Earth Modeling System (REMS) in order to be able to make assessments of the potential effects to changes in the climate and nutrient forcing regimes within which the coastal areas exist. The development path that OOSA follows is Observations Analysis Simulation Validation Prediction. The long-term goals are to develop an operational capability for coastal ecosystem health and sustainability.

Present Activities:

Coastal Ocean Modeling

Our modeling efforts are primarily focused on developing a Regional Earth Modeling System (REMS) to simulate land-ocean-atmosphere processes within the U.S. coastal regions. Our present modeling capability is a collaborative effort with UCLA, Rutgers and the Scripps Institution of Oceanography (SIO) and is developing a 3D coupled circulation/biogeochemical model for the eastern and western U.S. coasts. The circulation model is the Regional Ocean Modeling System (ROMS) developed at Rutgers/UCLA/SIO. The model domain ranges from the border of the U.S. and Canada on the north and the U.S. and Mexico on the south with nominal resolutions of 10 km. The biogeochemical model (Figure 1) is a complex multi-nutrient model that is able to track the pathways of oxygen, carbon, nitrogen, and phosphate. In addition to the 3D modeling efforts, we are also using 1D mixed-layer models and simpler box models to develop additional biogeochemical modeling components.

Figure 1. Schematic of the biogeochemical model. Not all the component pathways are shown. Carbon dioxide and oxygen are allowed across the air-sea and sediment-water boundaries. Benthic process pathways are not shown.

Recent results from this model for the west coast domain (Figure 2) show good agreement with the satellite-derived ocean color estimates for chlorophyll a. The one-to-one correlations of the upper ocean (0—100 m) values (not shown) are greater than 0.9 for SST, SSS and density values when compared against California Cooperative Fisheries Investigation (CalCOFI) historical data. Values for nutrients and chlorophyll drop to around 0.7 to 0.9—still a significant. Along the Southern California Bight region, a data assimilation effort (with SIO) is ongoing to develop both the capability to create the most reasonable initial conditions possible for carrying out real-time data assimilation of biogeochemical models. In addition, SIO is working with Rutgers to develop an MPI version of the ROMS code for use in the new NASA Compaq computer system.

Figure 2. Comparison between the observed (top panels) and the model-simulated (bottom panels) mean (left panels) and variance (right panels) fields for the May chlorophyll a values.

Because of the large influence that the Gulf Stream has along the eastern U.S. coasts, the efforts to develop a circulation model in this region have had to focus on developing a basin-scale model to resolve the Gulf Stream. This work is being carried out with Rutgers University. A smaller domain/grid has been developed for the North East North Atlantic (NENA) that uses 1-way nesting between the larger basin-scale domain and the smaller but more resolved coastal domain (Figure 3). Our present efforts are working to include watershed and air-shed fluxes for all of the model domains. Additionally, the East Coast model domain is being extended to include the entire Gulf of Mexico so that the Mississippi River outflow—a significant carbon pathway—can be resolved.

Coastal Observations

In the past year, we have released 9 Autonomous Drifting Ocean Sensor (ADOS; Figure 4) platforms along the West Coast of the U.S. The drifters are capable of measuring temperature profiles, surface downwelling irradiance and upwelling irradiance (in wavelength bands similar to those for SeaWiFS; Figure 5), surface atmospheric pressure, wind speed and direction. Additional deployments are planned for the summer of 2002. The data from these platforms is being used to estimate the spatial and temporal decorrelation scales for ocean color and the effects of wind events on ocean color variations.

Coastal Data Analysis

Our present data analysis efforts are focused on developing forcing data sets for supporting the model simulations presented above. These include: watershed and air-shed flux estimates; air-sea heat flux calculations; calculation of Eulerian and Lagrangian decorrelation scales of the satellite-derived SST and Ocean Color data sets. These decorrelation scales are required for use in objective mapping efforts and for comparison between model results and data. In addition, OOSA has been collaborating with Frank Hoge to investigate modeling of the CDOM climatologies extracted from the SeaWiFS observations, and with Steve Long (614) and Norden Huang (971) using EMD/HHT to investigate climate variations in chlorophyll a observations.

Figure 3: (top) Sea Surface Temperature (SST) image of the full domain of the North Atlantic (NATL) basin model. SST solution (bottom) is shown for day 1140 of the calculation using a 10 km horizontal grid resolution. The solid black line/boundary denotes the output stations for the 1-way nested open boundary conditions for the North-East North Atlantic (NENA) domain.

Educational Outreach

During the summer of 2001,we sponsored 2 students, a high school student and a Master’s Student. They worked on OOSA-related projects such as web development and ADOS drifter analysis. This coming summer, OOSA will sponsor 2 high school and 3 college students to work as summer interns.

Future/Planned Activities:

Coastal Land-Ocean-Atmosphere Simulations

We are working with Rutgers/UCLA/SIO to develop additional modeling capabilities. These include: a sediments transport model; storms surge; grid nesting; a regional atmospheric model; wet and dry deposition processes; a watershed model; and a more sophisticated benthic model capable of resolving processes from Submerged Aquatic Vegetation (SAV) and marshes. These additional components will allow us to develop the coupled circulation/biogeochemical ocean model into a complete Regional Earth Modeling System (REMS). This modeling package will then allow us to address many of the Earth Science Enterprise questions. We note here that our regional modeling efforts are resolving processes at scales comparable to those measured by present NASA remote sensing platforms (Less than 10 km). Present-day global-scale models are not expected to achieve this resolution capability for another 10-15 years. Additionally, J Moisan is working with T. Moisan (972), Pam Pittman (588) and Tony Baldwin (588) to develop a “Modeling Workbench” Facility. This facility will be used to investigate poorly resolved biogeochemical processes in order to develop a sophisticated biogeochemical model of phytoplankton growth.

Figure 4: Schematic of the ADOS platform (top) and a sample bio-optical data set (bottom) derived from the bio-optical sensors.
Figure 5: A schematic side view of OASIS. The blue feature along the hull is the solar panel array. The mast is designed to support a number of air-sea flux and weather sensors. The interior of the hull houses batteries, computers, communications devices and other environmental sensors. The entire platform is 2 feet in diameter and about 15 feet long.

Coastal Observations and Data Analysis

OOSA, in collaboration with Code 500 engineers (Pam Pittman and Troy Ames), is developing a low-cost Ocean-Atmosphere Sensor Integration System (OASIS; Figure 5). OASIS is a Surface Autonomous Vehicle (SAV) that is powered with solar panels, is capable of real-time two-way communication via an Iridium modem, is commanded and controlled using NASA’s Sensor Web Technology, and is designed to be low-cost (less than $20K) and support a wide array of air-sea physical and biogeochemical sensors—many of which are designed to support NASA Cal/Val efforts. A prototype OASIS has been funded by the Earth Science and Technology Office and plans are to commercialize it through SBIR and to carry out air-sea flux and bio-optical studies through National Ocean Partnership (NOPP) support—a proposal was submitted in April 2002. In addition to development of the OASIS platform, we anticipate leading in the development of the following facilities in support of coastal research. Several of these facilities would be designed as a general research facility to support participation from other NASA researchers and outside collaborators. These facilities include:

Coastal Ocean Radar Facility

We have been looking for a funding source to purchase a 2-site, long-range Seasonde network. The transponder/receiver sites would be located at Assateague Island, MD and Cape Henry, VA and would tie in with the Rutgers’ Cape May (DE) transponder/receiver to the north. The 3 units would allow us to obtain real-time data on surface currents and wave state for the entire Eastern Shore. The major science need for this data is to provide Cal/Val for the coastal circulation model. In addition to this we would like to investigate the possibility of using ocean color data with surface currents to estimate offshore transport of carbon. Some initial tests of this idea have been carried out using the U.S. West Coast model and showed that along the coast the magnitude of the total carbon flux was highly correlated with the product of the offshore surface velocity and the ocean color chlorophyll a estimate. This effort would occur in collaboration with T. Moisan and researchers at Horn Point and ODU. Our hope is to use these combination CODAR/satellite products to measure these and other fluxes. The velocity data would be also be provided to real-time coastal ocean managers such as NOAA and to our regional collaborators such as ODU, Rutgers, and Horn Point Labs. We also believe that additional research should be directed at WFF to work towards developing the nation’s next generation coastal ocean radars. WFF has the technical expertise to develop these instruments and is in a unique location to carry out validation studies.

Coastal Ocean Distributed Active Archive Center (CO.DAAC)

Because we have developed and are running a fully coupled circulation/biogeochemical model of the coastal ocean, it is important that we have access to the data necessary for carrying out model—data comparisons. These data are Local Area Coverage (LAC) data for temperature, ocean chlorophyll, sea surface height, etc. We have developed data—model comparison software that has been using the Global Area Coverage (GAC) data. However, because the model resolutions are now much finer than the GAC data, we have a need to carry out model-LAC comparisons. In order to do this we would like to develop a Coastal Ocean Distributed Active Archive Center. At present, only GAC data are available from single archive locations. LAC data, for instance the NOAA Coastwatch program, is typically parceled out to regional distribution/archive sites. We feel that a central NASA facility to archive and distribute these data would also be of benefit to the U.S. Coastal Ocean Observing programs now under development. Data management is one area of concern for this effort. NASA has a considerable history in providing satellite data to the community through the PODACC and DAAC archive sites. CO.DAAC would focus on coastal ocean data distribution—with a focus on LAC resolution data. Part of the CO.DAAC activities would also be to collect and distribute coastal land data sets for use in ocean-land process study and data analysis as well as to support development of watershed modeling capabilities. The CO.DAAC activity should collaborate with the coastal ocean color Pathfinder activities being led by Watson Gregg at GSFC and the NOAA CoastWatch effort and collaborate with both DAAC (GSFC) and PO.DAAC (JPL) activities.

GIS and Remote Sensing Analysis Laboratory

Another effort related to modeling is development of spatial data sets to force, initialize and validate models. Our efforts in developing watershed and air-shed flux forcing fields has required us to use Graphical Information System (GIS) tools. In addition to this, Frank Hoge is continuing to develop ocean color algorithms and techniques that are allowing us to predict coastal dissolved organic material concentrations. Once the coastal modeling effort include wetland and watershed model components we will require more GIS-based data sets for developing initial and forcing conditions as well as to carry out model—data inter-comparisons. Additionally, Steve Long (972) and Norden Huang (971) have been working with J. Moisan to analyze ocean color data using the Hilbert-Huang Transform modal decomposition techniques. HHT is a powerful tool for use in analyzing ocean color, SST and other satellite data sets. OSB is working to develop a computer laboratory facility that would focus on GIS coastal dataset development, coastal satellite data analysis studies, algorithm development. In addition to this, the laboratory would develop the use of the Distributed Oceanographic Data System (DODS), a data system intended to allow researchers transparent access to oceanographic data, stored in any of several different file formats, across the Internet [].

Educational Outreach

We will continue to host summer high school and college students under a variety of NASA funding source. In addition, we propose to establish a summer graduate-level biogeochemical/ecosystem modeling school aimed at introducing early to mid-level graduate students to coastal biogeochemical modeling efforts. The central goal is to allow a graduate student to have access to the state-of-the-art computational models and allow them to plan, develop and carry out several numerical experiments over the course of the summer program. At the end of the course the students would leave with model results that would be useful in their research topic. For instance, a zooplankton ecologist might develop and test a variety of foraging strategy models within a full 3D ecosystem model.


Lead Investigator: John Moisan