--David Harring (herring@ltpsun.gsfc.nasa.gov), EOS AM Science Outreach coordinator, Science Systems & Applications, Inc. (SSAI)
(Editor's Note: Adetailed introduction to the MOBY/MOCE effort was published in the Januar/February 1994 issue of The Earth Observer. To access that article, please refer to http://eospso.gsfc.nasa.gov/eos_oberv/1_2_94/ p17.html)
In its continuing support of NASA's Earth Observing System (EOS), NOAA's MOBY/MOCE Team conducted ship cruises over this past spring, summer, and fall off the southern coast of Lanai, HI, to refurbish its Marine Optical Buoy (MOBY) and to obtain Marine Optical Characterization Experiment (MOCE) data that were used to initialize the Japanese Ocean Color and Temperature Sensor (OCTS), and NASA's newly-launched Sea-viewing Wide Field-of-view Sensor (SeaWiFS). (See Figure 1.) Funded jointly by NASA and NOAA, the ongoing MOBY/MOCE efforts are primarily in support of the SeaWiFS and MODIS (Moderate Resolution Imaging Spectroradiometer, launching next year aboard EOS AM-1) instruments.
MOBY's primary purpose is to measure visible and infrared solar radiation entering and emanating from the ocean (see Figure 2). By monitoring variations in the reflected radiation, other quantities can be derived, such as the abundance of microscopic marine plants (phytoplankton). Over the last 3 months, the upgraded MOBY has provided an excellent time-series database facilitating SeaWiFS and MODIS bio-optical algorithm development for remote sensing over oligotrophic (low biomass productivity) ocean waters. Complementing the buoy data, a number of MOCE campaigns were also conducted to collect some 38 oceanic physical and bio-optical measurements such as radiometry, pigment analysis, total suspended matter, beam transmittance, and attenuation coefficients.
"To make global observations from space, there must be long-term independent checks to ensure that the accuracy is there," explains Dennis Clark, MOBY/MOCE leader. "Our basic measurements, like phytoplankton concentrations, will be used to validate SeaWiFS' and MODIS' global primary productivity data products, which in turn figure significantly into the global carbon balance."
"We are also looking at what causes changes in ocean color, which is a basic science question," adds Charles Trees, oceanographer at San Diego State University's Center for Hydro-optics and Remote Sensing (CHORS). "What are the processes governing ocean color change? We're also gaining insights into those processes."
Although its prototype instrumentation has been successfully collecting scientific data over the last three years, the team has continually worked hard to refine and refurbish its facilities beyond state-of-the-art. To support these ongoing upgrades to instruments and software, Clark has assembled an elite and complementary team of scientists, programmers, and engineers„all hand-picked for their rich and diverse professional backgrounds. The end result is a suite of measurement capabilities that, collectively, render the MOBY/MOCE Team rare to the point of being unique in all the world.
According to GSFC's Charles McClain, SeaWiFS Project Scientist and Cal/Val Manager, MOBY is the cornerstone of both the MODIS and SeaWiFS Teams' strategy for vicarious calibration of their ocean color detectors, and MOCE is one of the primary efforts for validation of their data products. Yet, with the delay of the SeaWiFS launch date to Aug. 1, 1997, and MODIS not scheduled to launch until June 30, 1998, the 1996 OCTS launch aboard the Japanese ADEOS platform provided the team its first opportunity to help initialize a satellite. (Unfortunately, ADEOS was rendered inoperable early this summer when its solar panel failed.) McClain explained that as part of the EOS theme of international cooperation and sharing data, the Japanese space agency (NASDA) relied on NOAA and NASA to provide infrastructure to help calibrate its sensors and validate their data products. In its first test as a satellite initialization station, the MOBY/MOCE operation proved to be a resounding success. Using software developed by GSFC's Watson Gregg and Gene Feldman (SeaWiFS Project members), NASA began receiving, processing, and distributing OCTS data in near real-time as early as October 1996. Shortly thereafter, Gregg and Feldman discovered some erroneous results in the OCTS data represented by negative water-leaving radiance measurements. "It became pretty clear that the instrument had fallen out of pre-launch calibration specifications," Gregg stated. "That's when we began looking for in situ data corresponding to overpass data. Dennis (Clark's MOBY/MOCE Team) provided us with high quality in situ data. By comparing the water-leaving radiances measured by OCTS to those measured at the surface, we were able to iteratively adjust the total at-satellite radiance. Then, after applying an atmospheric correction algorithm, I arrived at the new OCTS water-leaving radiance values."
Gregg says it is not known exactly how OCTS lost its pre-launch calibration. Perhaps there was degradation due to either (or both) vibration during launch, or out-gassing during deployment. Whatever the cause, shifts in pre-launch calibration are common during the mission lives of space-based remote sensors and, therefore, must be anticipated and corrected.
"The technique of acquiring MOBY/MOCE in situ data and comparing them to satellite data has now been mastered," Gregg observes, "and so we expect to be able to apply the same technique to SeaWiFS and MODIS in a fairly straightforward manner." (Refer to Figure 3.)
According to McClain, the earliest measurements from SeaWiFS look good but the main initialization activities for SeaWiFS will not take place until January 1998. He summarizes the MOBY/MOCE contributions: "Dennis has the most advanced shipboard system in the world at this point in terms of the suite of measurements he can make to support a satellite mission. He offers truly unique, first class facilities."
Aside from helping to initialize satellite sensors, both MOBY and MOCE have yielded some interesting scientific results on their own. For instance, due to recent technological upgrades in the buoy's optical system, giving it higher spectral resolution and greater sensitivity, the team is now able to see Fraunhofer lines in situ. Fraunhofer lines are constant, narrow gaps in the solar electromagnetic spectrum where the sun's atmosphere has absorbed energy. But, why is this significant?
"If we measure energy where there should be gaps (Fraunhofer lines), then that means energy is being scattered (Raman scattering) into the gaps from other parts of the spectrum," Clark reasons. "This improved observation capability is important in the remote sensing of fluorescence line height, or energy emitted by phytoplankton during its photosynthetic process, and enables more-accurate correction of those data. "
"This is also significant in that ordinarily it is not possible to get more energy at deeper ocean depths (due to absorption and reflection of incoming sunlight)," Clark adds, "but it could appear so in some wavelengths due to Raman scattering."
As part of its ongoing operations, the team has been working to evaluate and refine the SeaWiFS beta correction protocols to the measurement of reflection and absorption of light in the ocean. "There are classical protocols written for in-water radiometry and these are well understood," Trees explains. "Now, we are using a new, above-surface protocol for remote sensing of reflectance and evaluating the uncertainty between the two measurement protocols."
"Chuck Trees, through MOCE, has one of the best high-performance liquid chromatography (HPLC) data sets in the world," adds Clark. "By using this 12-year data set, we can compare systematic differences in old measurement techniques to his new HPLC technique using old data." With a better understanding of the errors inherent in the older validation measurement techniques, Clark hopes to pave the way for improved reprocessing of Coastal Zone Color Scanner (CZCS) data„a heritage instrument for both SeaWiFS and MODIS.
As another dividend from improvements in its measurement capabilities, the team recently learned that there is another aspect to pigment measurement gain which must account for the presence of divinyls„a form of chlorophyll a which is uniquely found in the pigment of picoplankton. "HPLC enables differentiation of compounds measured and enables us to improve our estimates," Trees states. "So, we have progressed scientifically to be able to get at real measurements of whatever compound you're interested in. For example, uncertainty in the measurements of CZCS data was about 25 percent, but now we're measuring all major pigment compounds to within 5-to-8 percent uncertainty."
Trees explains that the decrease in uncertainty is based on following the pigment protocols as outlined in the SeaWiFS Validation Plan and the U.S. JGOFS (Joint Global Ocean Flux Study) Plan. Extensive time and effort were spent in the separation and quantification of individual pigment compounds.
Clark and Trees hope to push the uncertainty figure even lower. "We are pushing the state of the art and refining our measurement techniques because we want the datasets from the next generation of remote sensors„SeaWiFS and MODIS„to be good for another 10-to-15 years. We want the best possible set of (validation) data to go into those measurements."
The team is also now obtaining very accurate solar sky radiance data across the spectrum at very high resolution, which will also be useful data to complement atmospheric modeling of the marine environment. To help in this effort, the MOBY/MOCE Team uses an AERONET CIMEL sun photometer on Lanai to measure atmospheric optical depth. This photometer is part of the worldwide AERONET (aerosol robotic network) deployed and maintained by GSFC's Brent Holben.
Mother nature offered the MOBY/MOCE Team a unique opportunity to observe the impact of atmospheric aerosols on regional measurements during the redeployment of MOBY this summer when the Pu'u O'o vent of Mt. Kilauea on the island of Hawaii became active again. Clark reports that, within a half-mile radius from the volcano's plume, the in-water light beam transmissivity over a 1-meter path length dropped by 15 to 20 percent. But, he adds, the measured chlorophyll fluorescence didn't change at all.
"We could see what appeared to be small, thin glass shards (silica) from the lava flow entering the water," Clark recalls. "Sometimes, there would be small explosions at the sea surface, sending some of these glass-like shards up into the atmosphere. We believe the decrease in transmissivity in the water is due to an increase in scattering from the silica shards."
Clark points out that, normally, the trade winds will carry particulates from the volcano away from MOBY; but, if there are wind shifts, it will help to know how these particulates affect the regional atmospheric properties.
What does it take to maintain a world-class marine optical measurement operation? Sometimes it takes engineers who are willing to work around the clock, under almost any circumstances, and who simply refuse to fail even when circumstances conspire against them.
Clark recounts an incident in which his team was due to make the delivery of an upgraded MOBY system to NASA in July of 1996. In January of that year, all of the new modifications were predicated upon upgrades to the spectrographs (modified charge-coupled devices), which had been delayed by the contractor. So, to help accelerate and assist in the delivery of the upgraded spectrographs, the MOBY team engineers went to Massachusetts and worked in the contractor's facility there to actually help with the assembly and testing of the prototypes. The net result of this upgrade is an improvement in measurement spectral resolution, while increasing the signal-to-noise ratio from 100-to-1 to 200-to-1.
"We discovered some major errors in the original mechanical design," Clark recalls, "so we redesigned and refabricated the new spectrographs at their facility. It was very nice of the contractor to give us that opportunity, but the bottom line is we wouldn't take ïno' for an answer. It had to be done."
No one can doubt the commitment of the divers who go down periodically to test the calibration of the currently deployed MOBY. A tough day at the office for them is swimming in rough Pacific waters to shine calibration lamps into the buoy's optics (see Figure 4) while 10-foot white tip sharks cruise by often to see what's on the menu. It is well known that larger fish, such as sharks or mahi, like to school around floating obstacles (e.g., buoys) in the water to prey upon the smaller fish that, in turn, prey upon whatever is growing on the obstacle. In short, a complete food chain is now flourishing around MOBY and its companion mooring buoy. On a number of occasions the divers have had to evacuate the waters when the sharks got too aggressive.
Through many changes since its prototype was built, MOBY has seemingly gone through a steady evolution. For instance, the material of the buoy itself was changed from a hollow aluminum cylinder to a solid nylon foam. Whereas the original design could have corroded and, if punctured, actually sunk, the new material corrodes much less readily and will not sink even if flooded. Also, the buoy's shape was changed to be much wider at the top and tapered toward the bottom, thereby further increasing its buoyancy.
Unfortunately, its increased buoyancy caused the new MOBY to move up and down in the water like a piston as it was rocked by waves. The team noticed that this motion added some "noise" to the data and also changed the depth dependency a little as MOBY's "arms" were being constantly shifted vertically. So, the MOBY team added fifteen anti-roll plates, suspended below the buoy at a depth of about 50 m, to increase drag and dampen the vertical oscillation of the buoy. Clark dubbed these plates "flopper stoppers." The flopper stoppers recently saved MOBY from drifting into oblivion when they reanchored MOBY on the shallow Penquin Banks off of Molokai, HI. MOBY was set adrift after its tether was cut, probably by the propeller of a passing fishing vessel.
Additionally, extra batteries and solar panels were added to increase the buoy's weight and prolong its capacity for operation between refurbishments. Stainless steel and copper anti-foulant tubes, with a granular bromide coating, were placed around the upwelling radiance collectors to limit the biofouling that the team had observed growing quickly over the optics of MOBY's earlier incarnations. Copper is widely used as a marine anti-foulant and has proven effective here, too. The steel and copper electrolytically interact, releasing a copper ion that reduces the fouling effects. The material comprising the irradiance collectors was also changed to a type of Teflon that is much more slippery and, therefore, reduces organisms' ability to attach themselves. Optically, this material was found to yield superior cosine response and increase throughput by 50-to-60 percent.
"MOBY is part of a calibration and validation infrastructure that is ready to go operational," states Clark. "We have also built an operations support infrastructure (at the MOBY/MOCE base in Honolulu) that is ready. We have added semi-clean rooms, a calibration lab, storage vans (acquired from the military), access to power, and a good communications infrastructure." (See Figure 5.) Good communications capabilities are essential for scientific field campaigns. The MOBY/MOCE Team has integrated its communications system with that of the University of Hawaii's satellite receiving station. This affords them access to both the local telephone service and to a T1 data line. The team also recently implemented an infrared relay system that enables fast communication from the MOBY mooring site, off the southwest coast of Lanai, to its base in the Honolulu shipyard. This has proven to be a major upgrade as the prototype system had trouble completing near-real-time cellular data relays, mainly due to the modems' burning up in the severe heat of the afternoon sun.
Today, not only MOBY's marine optical data, but also its housekeeping data are available in near real-time to team members as nearby as the lab site in Honolulu, or as far away as the Moss Landing Marine Lab in Monterey, CA. To take advantage of its new measurement and communications capabilities, Stephanie Flora, a programmer at Moss Landing, wrote new system software, tailor-made for the modified MOBY.
"When Stephanie arrives at work in the morning, MOBY has already called and sent its data to a computer in her lab, where she processes them and places them on a semi-secure Web page for the team's use," Clark explains. "Also, (the buoy's) diagnostic data are there, such as battery power or temperature, to characterize the housekeeping functions of the system. Ultimately, these data will be used by the SeaWiFS and MODIS teams to match with the data taken by those sensors."
Clark invites other SeaWiFS and MODIS team members to use the MOBY/MOCE facilities. "Very few other groups can accomplish what we can and still turn around data as quickly," he concludes. "We want this to be a world-class facility for supporting ocean remote sensing."
Mission accomplished.
