The Global Imager (GLI) was launched on board the Advanced Earth Observing Satellite II (ADEOS-II) on December 14, 2002. We conducted vicarious calibration of the GLI ocean color channels in visible to near-infrared channels. For the calibration we used the normalized water-leaving radiance derived from the Marine Optical Buoy (MOBY), and the aerosol optical properties (aerosol optical depth, size distribution, and refractive index) released in the Aerosol Robotic Network (AERONET). The following GLI characteristics are recognized from the calibration results. First, GLI underestimates the radiance in channels 1, 2, 4, and 5. Next, in near-infrared channels, it is suggested that GLI overestimates the radiance on the order of 15% in channels 18 and 19. Furthermore, the comparison of the result with other vicarious calibration results suggests the possibility that the GLI observed radiance has offset radiance versus the simulated radiance. The estimated offset is about 0.4 W/m²/um/sr in channel 19, which is considered appropriate by the adaptation test to the GLI standard atmospheric correction algorithm.

The multilayer perceptron (MLP) neural network have been widely used to fit non-linear transfer function and performed well. In this study, we use MLP to estimate chlorophyll-a concentrations from marine reflectance measures. The optical data were assembled from the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) Bio-optical Algorithm Mini-workshop (SeaBAM). Most bio-optical algorithms use simple ratios of reflectance in blue and green bands or combinations of ratios as parameters for regression analysis. Regression analysis has limitations for nonlinear function. Neural network, however, have been shown better performance for nonlinear problems. The result showed that accuracy of chlorophyll-a concentration using MLP is much higher than that of regression method. Nevertheless, using all of the five bands as input can derive the best performance. The results showed that each band could carry some useful messages for ocean color remote sensing. Only using band ratio (OC2) or band switch (OC4) might lose some available information. By preprocessing reflectance data with the principle component analysis (PCA), MLP could derive much better accuracy than traditional methods. The result showed that the reflectance of all bands should not be ignored for deriving the chlorophyll-a concentration because each band carries different useful ocean color information.

The remote sensing of ocean color, a problem that consists in retrieving one or several oceanic variables from top-of-atmosphere spectral reflectance, is considered as a collection of similar inverse problems continuously indexed by the angular variables influencing the observation process. A general solution is proposed in the form of a field of non-linear regression models over the set T of permitted values for the angular variables, i.e., as a map from T to some function space. Each value of the field is a regression model that performs a direct mapping from the top-of-atmosphere reflectance to the geophysical variable(s) of interest. A methodology based on ridge functions is developed to approximate this solution to an arbitrary accuracy, and is applied to the retrieval of the marine reflectance. The developed models are evaluated on synthetic data as well as on actual data originating from the SeaWiFS instrument. The retrievals are achieved with a good performance in terms of accuracy, robustness, and generalization capabilities, suggesting that the methodology might improve the inversion quality over existing techniques.

Recent work has shown the need for accurate treatment of radiative transfer in ocean color retrieval. The plane-parallel coupled atmosphere-ocean discrete ordinate model CAO-DISORT has been used to investigate the validity of current approximative inverse methods and to study new techniques for improved ocean color retrieval. In this paper we show that CAO-DISORT is fully differentiable with respect to its input optical properties, so that we can define analytic Jacobians with respect to any profile element in the atmosphere and ocean. A single call to the linearized model will produce radiances and Jacobians at arbitrary optical depth and viewing geometry in either medium. The model also has a pseudo-spherical treatment for solar beam attenuation in a curved atmosphere. The linearized model can be used directly in iterative least-squares retrievals requiring forward model simulations of backscatter measurements and their parameter derivatives; there is no need for approximations involving an atmospheric correction. We demonstrate the model's new capability by performing closed-loop least squares fitting to simultaneously retrieve the aerosol optical thickness and marine chlorophyll concentration from a set of 6 synthetic measurements at SeaWifs wavelengths.

This paper proposes scheme of absorptive aerosol correction for ocean color data of Global Imager (GLI). 380 nm band (GLI band 1) is used to detect absorptive aerosol, and to correct absorption. Spectral dependency of absorption is given empirically from GLI scenes with absorptive aerosol. It is built in GLI atmospheric correction. Satellite-derived water-leaving radiance is compared with in-situ data over East China Sea under presence of absorptive aerosol. Estimation error of water-leaving radiance is decreased from 79 to 49% at 380nm, though error still remained. The scheme applied to GLI data with absorptive aerosol, and it was confirmed that this atmospheric correction was effective.

SeaWiFS algorithms for estimating chlorophyll-a concentration (chl-a) from space were evaluated at the Rio de la Plata estuary and its influence area. Twenty-eight stations were explored during two surveys that took place in the 2003 austral spring (November and December). Surface water samples were collected for chl-a, plankton taxonomy, plankton flow cytometry, and absorption by CDOM and suspended particles. Radiometry was measured with both above-water SIMBAD and below-water Li-Cor PAR radiometers. SeaWiFS-derived chl-a values were obtained using NASA's OC4-V4 bio-optical algorithm. A systematic overestimation of chl-a by SeaWiFS was observed, especially in the complex estuarine waters, by 110% on average, and was largely attributed to the bio-optical algorithm, not the atmospheric correction scheme. The overestimation was most dramatic in two areas: the mouth of the Rio de la Plata and the external region of the continental shelf. In the Rio de la Plata estuary, large CDOM absorption seemed to explain the discrepancies. Interestingly, in the surface waters of the outer edge of the continental shelf a conspicuous change in the phytoplankton composition was detected at two contiguous stations. A complete lack of diatoms and dinoflagellates was reported at those stations, and cocoid forms (<2μm) dominated phytoplankton composition (99%). The flow cytometry data also exhibited some peculiar features, i.e., pronounced peaks in picoplankton (0.2-2μm) abundance. At one of the two stations, relatively large specific absorption by particles contributed to chl-a overestimation by the SeaWiFS bio-optical algorithm.

A simple, yet efficient and fairly accurate algorithm is presented to estimate photosynthetically available radiation (PAR) at the ocean surface from Global Imager (GLI) data. The algorithm utilizes plane-parallel radiation-transfer theory and separates the effects of the clear atmosphere and clouds. PAR is computed as the difference between the 400-700 nm solar flux incident at the top of the atmosphere (known) and reflected back to space by the atmosphere and surface (derived from GLI radiance), taking into account atmospheric absorption. Knowledge of pixel composition is not required, eliminating the need for cloud screening and arbitrary assumptions about sub-pixel cloudiness. For each GLI pixel, clear or cloudy, a daily PAR estimate is obtained. Diurnal changes in cloudiness are taken into account statistically, using a regional diurnal albedo climatology based on 5 years of ERBS data. The algorithm results are verified against other satellite estimates of PAR, the NCEP reanalysis product, and in-situ measurements from fixed buoys. Agreement is generally good between SeaWiFS and GLI estimates, with rms differences of 10 (28%), 5 (14%), and 3 (8.5%) Einstein/m2/day on daily, weekly, and monthly time scales. The rms differences between GLI and VISSR estimates and GLI and NCEP estimates are larger, i.e., 7.2 (15%) and 9 (30%) Einstein/m2/day, respectively, on monthly values. The comparison with buoy data also shows good agreement, with inaccuracies of 11.6 (28%), 10.3 (26%), and 7.1 (18%) Einstein/m2/day on daily, weekly, and monthly time scales. The good statistical performance makes the algorithm suitable for large-scale studies of aquatic photosynthesis.

Photo-thermo-plastic film (PTPF) is a multi-layer structure with the resolving power up to 1000 line pairs per millimeter in the binary and/or half-tone optical data recording modes. These structures are high-sensitive in the spectral range from 400 to 800 nm which is determined by chalcogenide glassy semiconductors (CGS) layer in the PTPF. We technologically challenged the CGS by tin-doping; this allows satisfying to main requirements which high-efficient observation systems are demanding. PTPF-based devices imply some critical elements for providing PTPF sensitization by means of the corona discharge as well as thermal development of the latent image to the form of superficial relief on the PTPF. Such PTPF-based slit camera was used for airborne monitoring of the Black Sea surface from the 9000-m-altitude. Camera resolving power is high enough for determining of waves heights and spacing as well for discovering of small sea objects and determining of their speed and drift direction. PTPF-based remote sensing seems to be even more advantageous due to the possibility to record different images multiple (up to 100) times on a single PTPF frame within the "recording - read-out - thermal erasing - re-recording" cycle. An algorithm for automatic measurements of the sea surface conditions is proposed. The measured parameters are height and spacing of waves as well as their motion direction. Mathematical processing includes 2-D smoothing of sample data, forming 1-D profile of the waves, and calculating its Fourier transform. By introducing of the scale factors, it makes possible to obtain certain data on the waves' characteristics. This system allows compressing of 2-D information to numerical data flow which is characterizing the rough seas and transmitting of these data through communication channels.

WindSat is a satellite-based multi-frequency polarimetric microwave radiometer designed to measure the fully polarimetric radiometric brightness temperature (TB) at 10.7, 18.7, and 37.0 GHz, and linearly polarized TB at 6.8 and 23.8 GHz. The primary goal of WindSat is to demonstrate the capability of polarimetric microwave radiometry in remote sensing of the ocean surface wind vector. Sea surface temperature, water vapor, and cloud liquid water, are among some of the other geophysical parameters that can also be measured. Solution of an inverse method to retrieve these environmental parameters from the polarimetric radiometer measurements requires a forward model that characterizes the measured brightness temperature due to emission from the surface and the intervening atmosphere, and its transmission through the atmosphere. This paper concentrates predominantly on the atmospheric component of the Windsat forward model. This includes two separate but related algorithms. We have developed a complete radiative transfer model that calculates the upward and downward atmospheric radiation, including attenuation effects due to clear air (both resonant and continuum absorption by water vapor and molecular oxygen) and non-precipitating clouds. From calculations and analysis of using this full forward model and an extensive match-up dataset a computationally efficient one-layer parameterized model has been developed for use in the physically-based Windsat retrieval algorithm. A performance assessment of forward models which utilize surface and atmospheric models, NWP model data assimilation profiles, and environmental measurements from other satellites, by comparison with WindSat measurements, is presented.

Using several months of WindSat measurements collocated with the NCEP Global Data Assimilation System model field, the Special Sensor Microwave Imager (SSM/I) measurements and QuikScat scatterometry measurements, we have derived an empirical geophysical model that describes radiometric vector for all WindSat channels, as a function of surface parameters: wind speed, wind direction and sea surface temperature, and atmospheric parameters: total precipitable water and cloud liquid water. This model function was then used to develop an ocean surface wind vector retrieval algorithm from WindSat polarimetric measurements. The accuracy of the retrieved wind vectors was quantified using several months of WindSat measurements collocated with the Special Sensor Microwave Imager (SSM/I) measurements and QuikSCAT scatterometry measurements.

Collocation of WindSat brightness temperatures and retrieved windspeeds with Stepped Frequency Microwave Radiometer (SFMR) on board NOAA WP-3D Orion aircraft are used to analyze and validate WindSat data under extreme wind conditions. For this study, the data presented are from September 2003 during Hurricane Fabian (03 September 2003). Temporal and spatial variability within the SFMR data stream are discussed in relation to both the WindSat brightness temperatures and derived windspeeds.

The Hyperspectral Imager for Coastal Oceans (HICO) sensor system, integrated in the International Space Station (ISS) Window Observational Research Facility (WORF), will collect visible and short-wave infrared hyperspectral data that will provide the following characterization of coastal regions: - Determine water clarity and visibility, shallow water bathymetry, and bottom type composition. - Detect underwater obstructions and characterize beaches and coastal areas. - Research global properties of coral reefs, the maritime atmosphere and determine global distribution of fires and active volcanoes in the context of mitigating natural hazards. It will achieve these objectives by collecting hyperspectral imaging data for over 70% of the Earth's surface, the portion flown over by ISS, at a spatial resolution of 25 meters. The desired data will be obtained using the Naval Research Lab (NRL) Portable Hyperspectral Imager for Low Light Spectroscopy (PHILLS-3) sensor with a pointing and stabilization system and then later integrating it with a short-wave infrared hyperspectral imager.

This paper studies the effect of internal wave in the Lombok Strait to chlorophyll distribution in the surrounded areas using ERS SAR, ASTER, SeaWiFS and AVHRR-NOAA images data during 1996-2004 periods. The observation results shows that the internal waves were propagated to the south and the north of strait and mostly occurred during transitional season from dry to wet and wet season (rainy season) between September to December when the layers are strongly stratified. Wavelet transform of image using Meyer wavelet analysis is applied for internal wave detection in ERS SAR and ASTER images, for symmetric extension of data at the image boundaries, to prevent discontinuities by a periodic wrapping of data in fast algorithm and space-saving code. Internal wave created elongated pattern in detail and approximation of image from level 2 to 5 and retained value between 2-4.59 times compared to sea surface, provided accuracy in classification over than 80%. In segmentation process, the Canny edge detector is applied on the approximation image at level two to derive internal wave signature in image. The proposed method can extract the internal wave signature, maintain the continuity of crest line while reduce small strikes from noise. The segmentation result, i.e. the length between crest and trough, is used to compute the internal wave induced current using Korteweg-de Vries (KdV) equation. On ERS SAR data contains surface signature of internal wave (2001/8/20), we calculated that internal wave propagation speed was 1.2 m/s and internal wave induced current was 0.56 m/s, respectively. From the observation of ERS SAR and SeaWiFS images data, we found out that the distribution of maximum chlorophyll area at southern coastline off Bali Island when strong internal wave induced current occurred in south of the Lombok Strait was distributed further to westward, i.e. from 9.25°-10.25°LS, 115°-116.25°SE to 8.8°-10.7°LS, 114.5°-116°SE, and surface chlorophyll concentration near coastal area, i.e. area 8.8°-9.25° LS, 114.5°-115°SE, increased. The preliminary result of this study concludes that the internal waves presumably affect chlorophyll distribution to westward (from 9.25°-10.25°LS, 115°-116.25°SE to 8.8°-10.7°LS, 114.5°-116°SE) in the south coast off Bali Island and increase surface chlorophyll concentration near coastal area (8.8°-9.25° LS, 114.5°-115°SE).

We observed sea surface signature of internal waves in ERS SAR and ASTER image over southwest of Japan during 193-2004 period. Wavelet transform has been applied for multi resolution of internal wave detection in image. Proposed method based on Meyer wavelet and Canny edge detector has been useful to delineate internal wave feature in ERS SAR and ASTER image, maintain the continuity of crest line while reduce small strikes from noise. Internal wave features were detected as elongated pattern in image with higher wavelet coefficient (>36) than sea surface (<10) on horizontal and vertical detail coefficient of image transforms at level 2-5, provided the accuracy in classification over than 80%. Spectral reflectance of internal wave signature was derived from ASTER image data. We found out that internal wave feature has higher reflectance at VNIR channel, between 520-660n, and lower reflectance at SWIR channel, between 760-2430nm, due to the absorption and scattering of material. The highest and the lowest reflectance were respectively 660nm and 760-810nm. The observation results show that the internal waves were mostly occurred at north coast off Kitakyushu and NW/W/SW/E coast off Tsushima Island on June to September period, during summer when the layers are strongly stratified. The length of the internal waves was detected between 6-28 km and wavelength between 120m-1.28km. The directions of propagation were varied between NW-SW at eastern channel and N-SW at western channel of Tsushima Strait. The areas where internal wave signatures were detected coincide with T/P data where sea level anomaly data steeply increased between 2-5 cm.

Subject: INTEVAC hybrid photomultiplier vacuum tube IPD-280 with 18 mm GaAsP photocathode, imaging electron optics, ion trap and 0.5, 1.0 diameter Schottky barrier anode. Problem: Large area intensified photodiodes (IPDs) have parameters (high sensitivity, gain, speed of operation, bandwidth, low noises), which are ideal for Ocean optic applications. However, these IPDs have not enough dynamic range and lifetime. Target of objective investigation: Identify the cause for small dynamic range and short lifetime of IPDs and optimize them for Ocean Optic applications. The voltages applied to photocathode and focusing electrodes have been experimentally optimized for maximal IPD sensitivity,dynamic range, pulse rise, and transit time. The photoelectrons trajectories and ions have been simulated using SIMION 3D 7,0 software for various voltages applied to the focusing electrodes. The uniformity of the photocathode has been tested to determine the impact of ions on the photocathode. Electron and ion currents investigations have been made for both negative and positive voltages applied to the ion trap electrode. Optimizing the regime for electron focusing and minimizing the ion current impact to photocathode was determines as result of the investigation. Reducing the voltages applied to photocathode and focusing the electrodes from 8 KV to 4-6 KV decreased the ion current. In this regime, the gain of IPD does not decrease significantly and the rise time and transit time of IPD remined practically the same.

This paper presents preliminary results of the 3D-wind retrieval algorithm developed for the multi-look airborne Doppler radar measurements using various forms of Kalman filters. The data was collected inside Hurricane Lili during NOAA's 2002 Atlantic Hurricane Ocean Winds Field Experiment with University of Massachusetts's newly developed Imaging Wind and Rain Airborne Profiler (IWRAP). Two forms of Adaptive Kalman filter are presented for 3-D wind retrieval. Simulations of different wind field and radar parameters are made to investigate the performance of the selected filter. Preliminary results of the actual 3-D wind estimates are then obtained and compared with simultaneous and independent wind vector measurements by GPS dropwindsondes, surface wind speed measurements by a microwave radiometer and flight level wind vector measurements.

This paper presents preliminary results of the 3D-wind retrieval algorithm developed for the multi-look airborne Doppler radar measurements using various forms of Kalman filters. The data was collected inside Hurricane Lili during NOAA's 2002 Atlantic Hurricane Ocean Winds Field Experiment with University of Massachusetts's newly developed Imaging Wind and Rain Airborne Profiler (IWRAP). Two forms of Adaptive Kalman filter are presented for 3-D wind retrieval. Simulations of different wind field and radar parameters are made to investigate the performance of the selected filter. Preliminary results of the actual 3-D wind estimates are then obtained and compared with simultaneous and independent wind vector measurements by GPS dropwindsondes, surface wind speed measurements by a microwave radiometer and flight level wind vector measurements.

Based on the comprehensive analysis of the geology, geomorphology, hydrology, and sediments conditions in the Jiulongjiang estuary and Xiamen bay, the water depth, the under water landform and geomorphic, the distribution of the scouring and depositing change are studied with the digitalized technique. The land use, tidal flat, suspending sediments and the flow profile of the flood and ebb tides are interpreted from the remote sensing images.

A spectral optimization algorithm (SOA) has been developed for processing satellite data in marine waters. The algorithm couples an atmospheric aerosol model with a detailed water-reflectance model to simultaneously retrieve both atmospheric and ocean color parameters. A key feature separating SOA from standard algorithms is the retrieval of the absorption coefficient of colored detrital material, particulate backscattering and chlorophyll concentration in absorbing atmospheres. The same parameters can also be retrieved in Case 2 waters where higher particulate backscattering contributes to the near-infrared (NIR) water-leaving reflectance. Results of chlorophyll concentration are presented in this study using SeaWiFS data obtained for the Chesapeake Bay and Middle Atlantic Bight. Manipulations to the NIR measurements of a degree similar to the magnitude of sensor noise and calibration uncertainty show that the chlorophyll retrievals are highly sensitive to the error in the measured NIR total reflectance.

When aerosol from the Hawaii volcano plume are present then the water leaving radiance derived from the SeaWifs satellite using the SeaDas algorithm is too large. It is shown that this problem may be due to the use of an incorrect aerosol model. Furthermore, it is shown that using longer wavelengths may be needed to solve the atmospheric correction problem under Hawaii volcanic aerosols.