COASTAL AND OCEANIC OPTICAL MEASUREMENTS II

AN IMPROVED T-R SYSTEM FOR FILTERPAD MEASUREMENT OF PHYTOPLANKTON SPECTRAL ABSORPTION

Hargreaves, Bruce R.¹; Lance, Veronica P.²; Vaillancourt, Robert D.²; Marra, John F.<sup>2

1</sup>Lehigh University 31 Williams Drive, Bethlehem, PA, 18015, United States; ²Lamont-Doherty Earth Observatory, Columbia University, 61 Route 9W, PO Box 1000, Palisades, New York, 10964-8000, United States

The Transmittance-Reflectance (T-R) method for measuring spectral absorption of phytoplankton and other suspended particles in natural waters was introduced by Tassan and Ferrari in 1995 and subsequently improved by these authors.Their improved version of the venerable “Quantitative Filterpad Technique (QFT)” has the potential to dramatically increase the accuracy of particle absorption but it has not been widely adopted. Presented here is a new instrument plus modifications of the published T-R method representing additional improvements (pQFT-TR). This portable QFT-TR system provides battery-powered, rapid, compact, and accurate measurement of particle spectral absorption over UV, VIS, and NIR wavelengths and should facilitate wider adoption and new applications of the T-R approach.The new T-R system (designed to work with a portable computer and 25mm diameter glass fiber filters) includes reflectance integrating sphere, fiber optic diode array spectrometer, xenon flash lamp, optical filters & standards, and 12-volt DC power.Procedures have been developed to create and test a universal calibration of the system while removing various inherent artifacts caused by fluorescence, stray light, detector nonlinearity, and spectrometer wavelength errors. The new system has recently been tested by making hundreds of underway measurements of samples collected during the Southern Ocean GASEX cruise (February-April 2008, NASA award NNX07AV23G).Pigment absorption of cruise samples was determined by subtracting bleached particle spectra from the corresponding unbleached spectra as recommended by Tassan and Ferrari (1995) but using bleach diluted in isotonic sodium sulfate.Analysis of SO-GASEX cruise samples showed extremely high correspondence (r² greater than 95%) between the red peak pigment absorption (a_ph675 nm) and extracted chlorophyll-a concentration (determined by chlorophyll-a fluorescence measurements after methanol extraction). Correction of fluorescence artifacts in both %T and %R data increased a_ph675 by 44-50% (highest with low sample loading) compared to uncorrected measurements, and yielded average chlorophyll-specific absorption at 675 nm of 0.023 m² mg^-1.

LIGHT ABSORPTION BY NATURAL AQUATIC PARTICLES IN THE NEAR-INFRARED (700 – 900 NM) SPECTRAL REGION.

Roettgers, Ruediger¹; Bracher, Astrid ²; Gehnke, Steffen¹; Schmitt, Bettina²; Wozniak, Slawomir<sup>3

1</sup>Institute for Coastal Research, GKSS Max Planck Strasse 1, Geesthacht, --, D-21502, Germany; ²Alfred Wegener Institute for Polar and Marine Research, Bremerhaven, Europe, D-27570, Germany; ³Marine Optics Laboratory, Institute of Oceanology , , Sopot, Europe, 81-712, Poland

Light absorption by aquatic particle in the near-infrared spectral region is often assumed to be negligible. So, any observed attenuation in this spectral region was supposed to be due to errors induced by scattering, and a general practice to correct for this scattering error in absorption measurements is to subtract the attenuation of these NIR wavelengths from all other wavelengths. We used the sensitive quantitative filter technique to measure particulate absorption but placed the filter inside an integrating sphere. The positioning of the filter inside the sphere does strongly improve the precision of the whole measurement by making any scattering error negligible. This allowed us to perform very sensitive measurements in the NIR region of 700 to 900 nm. Samples from different environments (open ocean, coastal waters and a river) and geographical regions (North Sea, Baltic Sea and the Atlantic Ocean) were examined. In nearly all cases a significant NIR particulate absorption could be observed that was as well visible when the samples had been bleached. In many case the NIR particulate absorption was high, reaching even 40 % of the absorption at the red chlorophyll maximum (672 nm). A good correlation of NIR particulate absorption with back-scattering in the same spectral region and with the concentration of total suspended matter was observed. We conclude that near-infrared particulate absorption mainly by detritus and minerals is not negligible (but might sometimes be very low), and that a subtraction of the measured attenuation in this region might lead to a large underestimation of the real absorption at shorter wavelength.

PHYTOPLANKTON DYNAMICS IN THE GULF OF MAINE: SECRETS REVEALED BY SEVEN YEARS OF HOURLY OPTICAL OBSERVATIONS ON A MOORED ARRAY

Roesler, Collin S.¹; Barnard, Andrew H.²; Pettigrew, Neal R.<sup>3

1</sup>University of Maine Darling Marine Center, 193 Clark's Cove Road, Walpole, ME, 04573, United States; ²WET Labs, Philomath, Oregon, 97370, United States; ³University of Maine, 206 Libby Hall, Orono, Maine, 04469, United States

July 2008 represents the 7- year anniversary of the deployment of the buoy array known as the Gulf of Maine Ocean Observing System (GoMOOS). In addition to meteorologic and hydrographic sensors, these buoys have been instrumented with two flavors of optical sensing packages: the "small" packages consist of chlorophyll fluorometers and 4-channel irradiance sensors at one or two depths, the "large" packages include additionally ac9s, backscattering sensors, CDOM fluorometers, and 7-channel upwelling radiance sensors. This data set has provided an unparalleled view of phytoplankton dynamics in the Gulf of Maine. Seasonal springtime blooms propagate from the southwestern part of the gulf, along the coast to the northeast, following thermal stratification patterns. Seasonal autumn blooms propagate to the southwest following cooling, increased winds and mixing, although the pattern of destratification is highly variable and multiple fall blooms can be observed. This pattern is robust except in the presence of anomalous springtime precipitation events that coincide with high river discharge due to springtime melt water runoff. During these years, phytoplankton respond to the gulf-wide salinity-driven stratification and blooms occur synoptically along the shelf, months earlier than is typical. These freshwater anomaly years have also been associated with the development of anomalously strong blooms of the toxic dinoflagellate Alexandrium fundyense and subsequent paralytic shellfish-poisoning events. Our ability to optically discern phytoplankton functional types, in addition to phytoplankton biomass, may provide a method for providing early warning for such events. The pattern of phytoplankton bloom development observed by the GoMOOS array is fundamentally different from that represented by ocean color remote sensing because of the coherence of blooms to CDOM-rich river discharge, which confounds the remote assessment of chlorophyll. In situ observation of CDOM provide the capability for deconvolving the remotely sensed chlorophyll and CDOM patterns.