Postprint (published version

Optical measurements including remote sensing provide a potential tool for the identification of 1 dominant phytoplankton groups and for monitoring spatial and temporal changes in biodiversity 2 in the upper ocean. We examine the application of an unsupervised hierarchical cluster analysis 3 to phytoplankton pigment data and spectra of the absorption coefficient and remote-sensing 4 reflectance with the aim of discriminating different phytoplankton assemblages in open ocean 5 environments under non-bloom conditions. This technique is applied to an optical and 6 phytoplankton pigment data set collected at several stations within the eastern Atlantic Ocean, 7where the surface total chlorophyll-a concentration (TChla) ranged from 0.11 to 0.62 mg m -3 . 8Stations were selected on the basis of significant differences in the ratios of the two most 9 dominant accessory pigments relative to TChla, as derived from High Performance Liquid 10 Chromatography (HPLC) analysis. The performance of cluster analysis applied to absorption and 11 remote-sensing spectra is evaluated by comparisons with the cluster partitioning of the 12 corresponding HPLC pigment data, in which the pigment-based clusters serve as a reference for 13 identifying different phytoplankton assemblages. Two indices, cophenetic and Rand, are utilized 14 in these comparisons to quantify the degree of similarity between pigment-based and optical-15 based clusters. The use of spectral derivative analysis for the optical data was also evaluated, and 16 sensitivity tests were conducted to determine the influence of parameters used in these 17 calculations (spectral range, smoothing filter size, band separation). The results of our analyses 18 indicate that the second derivative calculated from hyperspectral (1 nm resolution) data of the 19 phytoplankton absorption coefficient, a ph (λ), and remote-sensing reflectance, R rs (λ), provide 20 better discrimination of phytoplankton pigment assemblages than traditional multispectral band-21 ratios or ordinary (non-differentiated) hyperspectral data of absorption and remote-sensing 22 reflectance. The most useful spectral region for this discrimination extends generally from 23 3 wavelengths of about 425 -435 nm to wavelengths within the 495 -540 nm range, although in 24 the case of phytoplankton absorption data a broader spectral region can also provide satisfactory 25 results. 26 27 4

Abstract. Here we present results of the first comprehensive study of sulphur compounds and methane in the oligotrophic tropical western Pacific Ocean. The concentrations of dimethylsuphide (DMS), dimethylsulphoniopropionate (DMSP), dimethylsulphoxide (DMSO), and methane (CH4), as well as various phytoplankton marker pigments in the surface ocean were measured along a north–south transit from Japan to Australia in October 2009. DMS (0.9 nmol L−1), dissolved DMSP (DMSPd, 1.6 nmol L−1) and particulate DMSP (DMSPp, 2 nmol L−1) concentrations were generally low, while dissolved DMSO (DMSOd, 4.4 nmol L−1) and particulate DMSO (DMSOp, 11.5 nmol L−1) concentrations were comparably enhanced. Positive correlations were found between DMSO and DMSP as well as DMSP and DMSO with chlorophyll a, which suggests a similar source for both compounds. Similar phytoplankton groups were identified as being important for the DMSO and DMSP pool, thus, the same algae taxa might produce both DMSP and DMSO. In contrast, phytoplankton seemed to play only a minor role for the DMS distribution in the western Pacific Ocean. The observed DMSPp : DMSOp ratios were very low and seem to be characteristic of oligotrophic tropical waters representing the extreme endpoint of the global DMSPp : DMSOp ratio vs SST relationship. It is most likely that nutrient limitation and oxidative stress in the tropical western Pacific Ocean triggered enhanced DMSO production leading to an accumulation of DMSO in the sea surface. Positive correlations between DMSPd and CH4, as well as between DMSO (particulate and total) and CH4, were found along the transit. We conclude that DMSP and DMSO and/or their degradation products might serve as potential substrates for CH4 production in the oxic surface layer of the western Pacific Ocean.

In this work, we analyzed the color and texture of irises, ocular prostheses, and cosmetic colored contact lenses measured by means of a multispectral system, which provides the CIE L*a*b* colorimetric coordinates of a high resolution image pixel by pixel. The same subject, who has dark brown irises, participated in the measurement of all the contact lenses. The CIE L*a*b* colorimetric coordinates were analyzed to classify the samples into three major groups (brown, blue and green) using a new algorithm developed for this purpose. This classification allowed us to carry out a comparison of the color associated with each set of samples, using the corresponding color gamuts in the CIE L*a*b* color space. Furthermore, we analyzed the iris color reproduction achieved by prostheses and contact lenses in terms of CIEDE2000 color differences, and obtained closer results with prostheses. In addition, we performed an analysis of texture by means of the color spatial distribution of all samples. This was achieved by means of two statistical approaches: first order statistics of image histograms and second order statistics using co-occurrence matrices. The results suggest that the texture associated with real irises, ocular prostheses and colored contact lenses is very different. This study provides useful information about the color and texture of irises that may help to establish a strategy for improving the techniques used in the manufacturing process of prostheses and colored contact lenses to obtain a better and more realistic appearance.