Standard ocean chlorophyll-a (Chla) products from currently operational satellite sensors Moderate Resolution Imaging Spectroradiometer (MODIS) Aqua and Visible Infrared Imager Radiometer Suite (VIIRS) underestimate medium and high in situ Chla concentrations and have approximately 9% bias between each other in the California Current. By using the regional optimization approach of Kahru et al., we minimized the differences between satellite estimates and in situ match-ups as well as between estimates of the two satellite sensors and created improved empirical algorithms for both sensors. The regionally optimized Chla estimates from MODIS-Aqua and VIIRS have no bias between each other, have improved retrievals at medium to high in situ Chla, and can be merged to improve temporal frequency and spatial coverage and to extend the merged time series.

With the proliferation of digital cameras and automatic acquisition systems, scientists can acquire vast numbers of images for quantitative analysis. However, much image analysis is conducted manually, which is both time consuming and prone to error. As a result, valuable scientific data from many domains sit dormant in image libraries awaiting annotation. This work addresses one such domain: coral reef coverage estimation. In this setting, the goal, as defined by coral reef ecologists, is to determine the percentage of the reef surface covered by rock, sand, algae, and corals; it is often desirable to resolve these taxa at the genus level or below. This is challenging since the data exhibit significant within class variation, the borders between classes are complex, and the viewpoints and image quality vary. We introduce Moorea Labeled Corals, a large multi-year dataset with 400,000 expert annotations, to the computer vision community, and argue that this type of ecological data provides an excellent opportunity for performance benchmarking. We also propose a novel algorithm using texture and color descriptors over multiple scales that outperforms commonly used techniques from the texture classification literature. We show that the proposed algorithm accurately estimates coral coverage across locations and years, thereby taking a significant step towards reliable automated coral reef image annotation.

After ADEOS-2 launch on 14 December 2002, GLI ocean group has investigated vicarious calibration, algorithm improvement and product validation, and released version-1 products on 24 December 2003. Although we lost ADEOS-2/GLI on 24 October 2003, we continue product evaluation and algorithm revision toward the future product updates. For example, we are investigating absorptive aerosol correction and 250-m ocean color image processing. The absorptive aerosol correction is very important especially in the East Asia where is surrounded by the fast developing countries and influenced frequently by a kind of carbon aerosols. The 250-m products are made using GLI 250-m channels from blue to near-infrared, and enable to detect fine spatial structures in the coastal areas.

An analysis of ocean color temperature scanner [(OCTS) on board the advanced earth observation satellite (ADEOS)] spectral radiance data was performed for retrieving global distributions of Angstrom factor and exponent, which represent the aerosol optical thickness at a reference wavelength (500 nm in the authors' study) and a spectral dependence of the optical thicknesses, respectively, over ocean. Determination of calibration constants for OCTS-received radiances and development of an efficient look-up table method for synthesizing the radiances are key issues for development of the present two-channel algorithm with use of channel 6 and 8 radiances of OCTS. This algorithm has been applied to Level-1B OCTS GAC data sets for producing three month (April, May, and June 1997) global distributions of Angstrom parameters. Geographical and seasonal distribution patterns of Angstrom parameters suggest that anthropogenic sulfate aerosols in mid-latitudes and biomass burning aerosols in the subtropical region are characterized by small particles having large Angstrom exponents, whereas mineral dust particles from subtropical arid regions are characterized by large particles having small Angstrom exponents. There was a fairly good agreement between satellite-retrieved values of Angstrom parameters and values obtained by sky radiometers located on coasts.