Feature ArticlesAdvanced Instrumentation For Optically Complex Waters
By John H. Morrow
Chief Executive Officer and Director
Biospherical Instruments Inc.
San Diego, California
Stanford B. Hooker
NASA Calibration and Validation
Oceanographic optical profiling systems often do not properly resolve the vertical complexity of shallow, nearshore waters because of their size, rate of descent or deployment protocols. With the goal of achieving greater radiometric accuracy at reduced costs, new aquatic optical profilers have recently been developed by Biospherical Instruments Inc. (BSI) in collaboration with researchers from the NASA Calibration and Validation Office.
NASA has a continuing requirement to collect ground-truth—more properly, sea-truth—observations for the vicarious calibration of ocean color satellite sensors and validation of the algorithms for which the remotely sensed observations are used as input parameters. The emphasis on nearshore processes in the Ocean Biology and Biogeochemistry Advanced Science Plan and corresponding satellite mission design concepts requires an unprecedented capability to collect high-quality data in shallow and optically complex waters. Within this context, 'high quality' refers to measurements with a documented uncertainty in keeping with established performance metrics for producing climate-quality data records.
Working with NASA, BSI has taken a new approach to meeting the need for an optical profiler optimized for optically complex waters. The resulting instrument, called the Compact-Optical Profiling System (C-OPS), uses high-speed microradiometer light sensors to produce 19 waveband radiance and irradiance instruments within a 6.4-centimeter housing—29 percent smaller than typical legacy instruments.
Developed by BSI as the 'next generation' photodetector, microradiometers are stand-alone, miniaturized, individual optical data acquisition systems that may be networked into complex instrument systems. In addition to improvements in the optical instrumentation, C-OPS benefits from a novel, kite-shaped free-fall backplane with a unique adaptive-buoyancy system that affords profiling speeds from five to 100 centimeters per second. Deployed by hand from vessels small and large, the kite-shaped system may be used to collect profiles of irradiance and radiance in waters from two to 350 meters' depth.
C-OPS' typical platform angle control is within 2° from vertical. Sampling speeds of 15 data frames per second covering 10 decades of dynamic range ensure that the data set is representative of the shallowest of waters. These next-generation instruments were specifically designed to deploy from small vessels—by hand—and to support multidisciplinary research, such as remote sensing of ocean color, phytoplankton ecology and distribution, harmful algal blooms, and the global carbon budget.
The benefits of this new sampling capability are expected to be lower uncertainties in the data products across the full dynamic range of the sampling problem set, better accuracy in separating the living and nonliving components of seawater, and an improved understanding of the interaction between the ocean and atmosphere.
Furthermore, the microradiometer technology is more easily expanded than legacy systems, ensuring a cost-effective expansion path for both apparent optical property profiling instruments and novel systems occupying new roles in the future. This activity is an important initial step toward the ability to support a coupled ocean-atmosphere observing system (i.e., a calibration and validation capability for a combined satellite mission). Such a mission will very likely emphasize coastal and open-ocean processes, so a capability for making high-quality measurements with equal ease in both the nearshore and open-ocean environments is an inevitable requirement.
There are four essential factors that must be optimized to acquire high-accuracy vertical profiles of the in-situ light field: instrument size, spectral resolution, sampling rate and deployment speed. C-OPS represents significant innovations in all of these areas.
The principal innovation affecting all design factors is the use of microradiometers wherever standard filter-photodetector packages have been used in the past. Their compact size ensures that a large number of optical sensors can be packaged in a small-diameter housing (e.g., 19 inches, seven centimeters) and that important spectral features of the water column are adequately sampled. The high sampling speeds (15 data frames per second) and wide dynamic range (10 decades) afforded by microradiometer technology are unprecedented in submersible profilers, where many legacy systems typically operate less than 50 percent as quickly and with less than 65 percent of the spectral resolution. The small diameter of the individual sensors and instrument package minimizes the shadow cast by the instrument during a vertical profile. Furthermore, it has long been known that any size boat is a significant perturbation to the light field (from shadows and reflections) and flow field (near-surface layers are significantly mixed).
Legacy free-fall systems that can be floated to a spot outside the influence of the boat all permit a sampling of the unperturbed natural environment. Building on this approach, C-OPS is well-suited to sample optically complex waters because the kite-shaped free-fall system houses a hydrobaric buoyancy chamber containing up to three air-filled bladders. Using syntactic foam flotation, the system is tuned to loiter near the surface, sampling continuously. As the system descends, water pressure compresses the bladders, reducing buoyancy and increasing the rate of descent until terminal velocity is reached.
Surface loitering, deep terminal velocity depth and high data rates result in sufficient sampling to reveal optically diverse, near-surface thin layers and statistically relevant data on surface effects (i.e., reduction of aliasing from wave focusing).
The planar geometry of the sensor apertures can be maintained using two quickly applied adjustments: moving buoyancy elements along the long axis of the deployment package to trim the roll of the system and applying a tilt offset to each sensor to overcome cable and current biases, which act along the short axis of the package. C-OPS can be tuned to avoid data degradation; it also can sample the natural environment better than prior instruments. Therefore, the presence of a previously challenging natural phenomenon—for example, wave-focusing effects or near-surface optically different layers—can be better handled in the data-processing scheme used to derive products from the light measurements, because they are adequately sampled.
The culmination of the advancements applied to the design of C-OPS can be seen in the quality of the data collected during field sampling from open ocean (deep, clear water) to river plumes (shallow, turbid water).
Remote sensing reflectance is the primary variable used in deriving ocean color data products such as chlorophyll a concentration from satellite observations.
Many of these algorithms are based on open-ocean relationships, wherein the maximum reflectance is in the blue-green part of the spectrum and there is almost no signal in the near-infrared. Preliminary C-OPS data from a transect during the recent Malina campaign to the Arctic show significant signals in the red and near-infrared, as well as dramatic differences between offshore—with and without ice—and inshore waters. Although not as notable as the red and near-infrared changes, the offshore to onshore differences are also seen in the ultraviolet portion of the spectrum.
Perhaps most importantly, data products from this transect are being produced across a very large range of wavelengths: 320 to 780 nanometers. This is made possible by the tunable descent velocity of the C-OPS profiler, which allows adequate data to be collected in waters where the attenuation is very large (i.e., the red and near-infrared in clear waters, as well as the ultraviolet in turbid waters.)
John H. Morrow came to Biospherical Instruments (BSI) in 1987 as a development engineer and biological oceanographer from the University of Southern California. Since then, he has been actively involved in the development and marketing of hand-deployed profilers. He has been president of BSI since 1998.
C.R. Booth was a young researcher in the Visibility Laboratory at the Scripps Institution of Oceanography when he started Biospherical Instruments (BSI) in 1977. Since then, BSI has produced a wide variety of optical instruments for the oceanographic and atmospheric science communities. Currently, he is the chief executive officer and director of research.
Stanford B. Hooker joined NASA to work on the Sea-Viewing Wide Field-of-View Sensor Project in 1991 and was the deputy project scientist when the mission office closed in 2004. He is currently the director of the Calibration and Validation Office for the ocean biology and biogeochemistry program at NASA headquarters.