Export 105 results:
Sort by: Author Title Type [ Year  (Desc)]
Lee, YJ, Matrai PA, Friedrichs MAM, Saba VS, Aumont O, Babin M, Buitenhuis ET, Chevallier M, de Mora L, Dessert M, Dunne JP, Ellingsen IH, Feldman D, Frouin R, Gehlen M, Gorgues T, Ilyina T, Jin MB, John JG, Lawrence J, Manizza M, Menkes CE, Perruche C, Le Fouest V, Popova EE, Romanou A, Samuelsen A, Schwinger J, Seferian R, Stock CA, Tjiputra J, Tremblay B, Ueyoshi K, Vichi M, Yool A, Zhang JL.  2016.  Net primary productivity estimates and environmental variables in the Arctic Ocean: An assessment of coupled physical-biogeochemical models. Journal of Geophysical Research-Oceans. 121:8635-8669.   10.1002/2016jc011993   AbstractWebsite

The relative skill of 21 regional and global biogeochemical models was assessed in terms of how well the models reproduced observed net primary productivity (NPP) and environmental variables such as nitrate concentration (NO3), mixed layer depth (MLD), euphotic layer depth (Z(eu)), and sea ice concentration, by comparing results against a newly updated, quality-controlled in situ NPP database for the Arctic Ocean (1959-2011). The models broadly captured the spatial features of integrated NPP (iNPP) on a pan-Arctic scale. Most models underestimated iNPP by varying degrees in spite of overestimating surface NO3, MLD, and Z(eu) throughout the regions. Among the models, iNPP exhibited little difference over sea ice condition (ice-free versus ice-influenced) and bottom depth (shelf versus deep ocean). The models performed relatively well for the most recent decade and toward the end of Arctic summer. In the Barents and Greenland Seas, regional model skill of surface NO3 was best associated with how well MLD was reproduced. Regionally, iNPP was relatively well simulated in the Beaufort Sea and the central Arctic Basin, where in situ NPP is low and nutrients are mostly depleted. Models performed less well at simulating iNPP in the Greenland and Chukchi Seas, despite the higher model skill in MLD and sea ice concentration, respectively. iNPP model skill was constrained by different factors in different Arctic Ocean regions. Our study suggests that better parameterization of biological and ecological microbial rates (phytoplankton growth and zooplankton grazing) are needed for improved Arctic Ocean biogeochemical modeling.

Laliberte, J, Belanger S, Frouin R.  2016.  Evaluation of satellite-based algorithms to estimate photosynthetically available radiation (PAR) reaching the ocean surface at high northern latitudes. Remote Sensing of Environment. 184:199-211.   10.1016/j.rse.2016.06.014   AbstractWebsite

Two satellite-based methods to estimate daily averaged photosynthetically available radiation (PAR) at the ocean surface are evaluated at high northern latitudes. The first method employs a precomputed Look-Up-Table (LUT) generated from radiative transfer simulations. The LUT associates spectral irradiance reaching the surface to a given set of input parameters, namely solar zenith angle, cloud optical thickness, cloud fraction, ozone concentration, and surface albedo. The second method, as implemented by NASA's Ocean Biology Processing Group (OBPG) in the standard Ocean Color data processing chain, expresses the energy budget of the atmosphere-surface-ocean system via a simple radiative transfer model. The performance of these methods is evaluated using an extensive in situ PAR dataset collected in the Arctic Ocean between 1998 and 2014, with daily values ranging from 0.08 to 61.07 Em(-2) d(-1). A methodology is developed to compare in situ measurements and satellite products of different spatial and temporal resolution. Agreement is generally good between satellite-derived estimates and ship-based data and between methods. Specifically, both methods yield a small positive bias of 6% and 2% and a relative uncertainty larger than that observed at low latitude, with a root mean squared error (RMSE) of 33% and 20% for the LUT and OPBG methods, respectively. This is attributed to the peculiar environmental conditions encountered in the Arctic, namely low solar elevation, changing surface albedo due to sea ice, and persistent cloudiness. The RMSE difference among methods is due to the high temporal resolution (3 h) of the International Satellite Cloud Climatology Project (ISCCP) LUT input not fully compensating for its low spatial resolution (280 km grid cells). The LUT method has the major advantage of providing PAR estimates in all conditions, including ice-covered regions, while the OBPG method is currently limited to open waters and a solar zenith angle lower than 83 degrees. Consequently, the OBPG method may not account for as much as 38% of PAR reaching the Arctic ocean surface annually. Both methods have the potential to provide useful PAR estimates just below the ice, by including information about ice transmittance. (C) 2016 Elsevier Inc. All rights reserved.

Lefevre, J, Menkes C, Bani P, Marchesiello P, Curci G, Grell GA, Frouin R.  2016.  Distribution of sulfur aerosol precursors in the SPCZ released by continuous volcanic degassing at Ambrym, Vanuatu. Journal of Volcanology and Geothermal Research. 322:76-104.   10.1016/j.jvolgeores.2015.07.018   AbstractWebsite

The Melanesian Volcanic Arc (MVA) emits about 12 kT d(-1) of sulfur dioxide (SO2) to the atmosphere from continuous passive (non-explosive) volcanic degassing, which contributes 20% of the global SO2 emission from volcanoes. Here we assess, from up-to-date and long-term observations, the SO2 emission of the Ambrym volcano, one of the dominant volcanoes in the MVA, and we investigate its role as sulfate precursor on the regional distribution of aerosols, using both satellite observations and model results at 1 x 1 spatial resolution from WRF-Chem/GOCART. Without considering aerosol forcing on clouds, our model parameterizations for convection, vertical mixing and cloud properties provide a reliable chemical weather representation, making possible a cross-examination of model solution and observations. This preliminary work enables the identification of biases and limitations affecting both the model (missing sources) and satellite sensors and algorithms (for aerosol detection and classification) and leads to the implementation of improved transport and aerosol processes in the modeling system. On the one hand, the model confirms a 50% underestimation of SO2 emissions due to satellite swath sampling of the Ozone Monitoring Instrument (OMI), consistent with field studies. The OMI irregular sampling also produces a level of noise that impairs its monitoring capacity during short-term volcanic events. On the other hand, the model reveals a large sensitivity on aerosol composition and Aerosol Optical Depth (AOD) due to choices of both the source function in WRF-Chem and size parameters for sea-salt in FIexAOD, the post-processor used to compute offline the simulated AOD. We then proceed to diagnosing the role of SO2 volcanic emission in the regional aerosol composition. The model shows that both dynamics and cloud properties associated with the South Pacific Convergence Zone (SPCZ) have a large influence on the oxidation of SO2 and on the transport pathways of volcanic species across the South Pacific atmosphere. For example, in the tropical cloudy air, the sulfate production in the aqueous phase is very efficient, resulting in the formation of a large cloud of highly scattering sulfate aerosols advected horizontally to Eastern Indonesia, in agreement with the AOD feature captured by MODIS/Aqua, but missed in CALIOP/CALIPSO (lidar) products. Model sensitivity experiments indicate that aerosol re-suspension due to evaporating droplets is a significant pathway for the supply of volcanic sulfur species in the remote marine boundary layer. By strongly modulating the irreversible loss due to wet scavenging, this aerosol process has a similar influence on the sulfur burden as natural emission from volcanoes or biogenic sources like dimethyl sulfate (DMS). The results emphasize the importance of MVA passive degassing and SPCZ dynamics on the aerosol background, and raise questions about potential impacts on the local climate and marine ecosystems. (C) 2015 Elsevier B.V. All rights reserved.

Valente, A, Sathyendranath S, Brotas V, Groom S, Grant M, Taberner M, Antoine D, Arnone R, Balch WM, Barker K, Barlow R, Belanger S, Berthon JF, Besiktepe S, Brando V, Canuti E, Chavez F, Claustre H, Crout R, Frouin R, Garcia-Soto C, Gibb S, Gould R, Hooker S, Kahru M, Klein H, Kratzer S, Loisel H, McKee D, Mitchell BG, Moisan T, Muller-Karger F, O'Dowd L, Ondrusek M, Poulton AJ, Repecaud M, Smyth T, Sosik HM, Twardowski M, Voss K, Werdell J, Wernand M, Zibordi G.  2016.  A compilation of global bio-optical in situ data for ocean-colour satellite applications. Earth System Science Data. 8:235-252.   10.5194/essd-8-235-2016   AbstractWebsite

A compiled set of in situ data is important to evaluate the quality of ocean-colour satellite-data records. Here we describe the data compiled for the validation of the ocean-colour products from the ESA Ocean Colour Climate Change Initiative (OC-CCI). The data were acquired from several sources (MOBY, BOUSSOLE, AERONET-OC, SeaBASS, NOMAD, MERMAID, AMT, ICES, HOT, GeP&CO), span between 1997 and 2012, and have a global distribution. Observations of the following variables were compiled: spectral remote-sensing reflectances, concentrations of chlorophyll a, spectral inherent optical properties and spectral diffuse attenuation coefficients. The data were from multi-project archives acquired via the open internet services or from individual projects, acquired directly from data providers. Methodologies were implemented for homogenisation, quality control and merging of all data. No changes were made to the original data, other than averaging of observations that were close in time and space, elimination of some points after quality control and conversion to a standard format. The final result is a merged table designed for validation of satellite-derived ocean-colour products and available in text format. Metadata of each in situ measurement (original source, cruise or experiment, principal investigator) were preserved throughout the work and made available in the final table. Using all the data in a validation exercise increases the number of matchups and enhances the representativeness of different marine regimes. By making available the metadata, it is also possible to analyse each set of data separately. The compiled data are available at doi: 10.1594/PANGAEA.854832 (Valente et al., 2015).

Lutz, V, Frouin R, Negri R, Silva R, Pompeu M, Rudorff N, Cabral A, Dogliotti A, Martinez G.  2016.  Bio-optical characteristics along the Straits of Magallanes. Continental Shelf Research. 119:56-67.   10.1016/j.csr.2016.03.008   AbstractWebsite

The Straits of Magallanes at the tip of South America connects the Atlantic and Pacific Oceans. The variability in the absorption characteristics by phytoplankton (a(ph)(440)), non-pigmented particles, NPP (a(NPP)(440)), and chromophoric dissolved organic matter, CDOM (a(y)(440)), measured along the Straits in late summer 2011 (RN Melville MV1102 cruise), was analyzed. Satellite-derived monthly PAR data showed that at the time of the cruise the western sector was exposed to a low-light environment (similar to 16 mol quanta m(-2)d(-1)) while the eastern sector received higher irradiance (similar to 28 mol quanta m(-2)d(-1)). In the Patagonian Shelf total absorption was dominated by phytoplankton (up to 76%; aph (440)=0.265 m(-1)), while in the Atlantic Sector of the Straits, the major contributor was NPP (up to 42%; a(NPP)(440)=0.138 m(-1)), and in the Pacific Sector of the Straits CDOM contributed up to 80% of the total absorption (a(y)(440)=0.232 m(-1)). These changes could be related in part to the input of fresh water from glacier melting and rain in the Pacific Sector (a(y)(440) vs salinity r(s)=-0.98). The carbon biomass (C) was composed in its majority by pico-phytoplanlcton and secondly by nano-phytoplankton, with exception of the Atlantic Sector where the micro-phytoplankton dominated. Carbon to chlorophyll-a ratios (C:Chla) were very low throughout the Straits (average of similar to 6) because of photoacclimation to the extremely low light Complementary pigments data obtained in spring 2003 by the BEAGLE expedition indicated the predominance of diatoms all along the Straits, but the bio-optical trend resembled the one found in late summer 2011, i.e., NPP dominated the absorption in the well mixed Atlantic Sector, phytoplankton in the Middle Sector, and CDOM in the Pacific Sector. These results emphasize that underwater light is the major factor affecting phytoplankton growth and physiology, and that prevalent physical and geochemical conditions play an important role regulating the bio-optical properties in this heterogeneous area. These effects should be considered to adjust parameters (such as C:Chla) when running biogeochemical models for this region. (C) 2016 Elsevier Ltd. All rights reserved.

Wattelez, G, Dupouy C, Mangeas M, Lefevre J, Touraivane, Frouin R.  2016.  A statistical algorithm for estimating chlorophyll concentration in the New Caledonian lagoon. Remote Sensing. 8   10.3390/rs8010045   AbstractWebsite

Spatial and temporal dynamics of phytoplankton biomass and water turbidity can provide crucial information about the function, health and vulnerability of lagoon ecosystems (coral reefs, sea grasses, etc.). A statistical algorithm is proposed to estimate chlorophyll-a concentration ([chl-a]) in optically complex waters of the New Caledonian lagoon from MODIS-derived remote-sensing reflectance (R-rs). The algorithm is developed via supervised learning on match-ups gathered from 2002 to 2010. The best performance is obtained by combining two models, selected according to the ratio of R-rs in spectral bands centered on 488 and 555 nm: a log-linear model for low [chl-a] (AFLC) and a support vector machine (SVM) model or a classic model (OC3) for high [chl-a]. The log-linear model is developed based on SVM regression analysis. This approach outperforms the classical OC3 approach, especially in shallow waters, with a root mean squared error 30% lower. The proposed algorithm enables more accurate assessments of [chl-a] and its variability in this typical oligo- to meso-trophic tropical lagoon, from shallow coastal waters and nearby reefs to deeper waters and in the open ocean.

Frouin, R, Pelletier B.  2015.  Bayesian methodology for inverting satellite ocean-color data. Remote Sensing of Environment. 159:332-360.   10.1016/j.rse.2014.12.001   AbstractWebsite

The inverse ocean color problem, i.e., the retrieval of marine reflectance from top-of-atmosphere (TOA) reflectance, is examined in a Bayesian context. The solution is expressed as a probability distribution that measures the likelihood of encountering specific values of the marine reflectance given the observed TOA reflectance. This conditional distribution, the posterior distribution, allows the construction of reliable multi-dimensional confidence domains of the retrieved marine reflectance. The expectation and covariance of the posterior distribution are computed, which gives for each pixel an estimate of the marine reflectance and a measure of its uncertainty. Situations for which forward model and observation are incompatible are also identified. Prior distributions of the forward model parameters that are suitable for use at the global scale, as well as a noise model, are determined. Partition-based models are defined and implemented for SeaWiFS, to approximate numerically the expectation and covariance. The ill-posed nature of the inverse problem is illustrated, indicating that a large set of ocean and atmospheric states, or pre-images, may correspond to very close values of the satellite signal. Theoretical performance is good globally, i.e., on average over all the geometric and geophysical situations considered, with negligible biases and standard deviation decreasing from 0.004 at 412 nm to 0.001 at 670 nm. Errors are smaller for geometries that avoid Sun glint and minimize air mass and aerosol influence, and for small aerosol optical thickness and maritime aerosols. The estimated uncertainty is consistent with the inversion error. The theoretical concepts and inverse models are applied to actual SeaWiFS imagery, and comparisons are made with estimates from the SeaDAS standard atmospheric correction algorithm and in situ measurements. The Bayesian and SeaDAS marine reflectance fields exhibit resemblance in patterns of variability, but the Bayesian imagery is less noisy and characterized by different spatial de-correlation scales. Experimental errors obtained from match-up data are similar to the theoretical errors determined from simulated data. Regionalization of the inverse models is a natural development to improve retrieval accuracy, for example by including explicit knowledge of the space and time variability of atmospheric variables. (C) 2014 Elsevier Inc. All rights reserved.

Belanger, S, Frouin R, Wang M, Goyens C, Stamnes K.  2015.  From surface to top-of-atmosphere. Ocean Colour Remote Sensing in Polar Seas. 16( Babin M, Arrigo K, Belanger S, Forget MH, Eds.).:27-59. Abstract
Zoffoli, ML, Frouin R, Kampel M.  2014.  Water column correction for coral reef studies by remote sensing. Sensors. 14:16881-16931.   10.3390/s140916881   AbstractWebsite

Human activity and natural climate trends constitute a major threat to coral reefs worldwide. Models predict a significant reduction in reef spatial extension together with a decline in biodiversity in the relatively near future. In this context, monitoring programs to detect changes in reef ecosystems are essential. In recent years, coral reef mapping using remote sensing data has benefited from instruments with better resolution and computational advances in storage and processing capabilities. However, the water column represents an additional complexity when extracting information from submerged substrates by remote sensing that demands a correction of its effect. In this article, the basic concepts of bottom substrate remote sensing and water column interference are presented. A compendium of methodologies developed to reduce water column effects in coral ecosystems studied by remote sensing that include their salient features, advantages and drawbacks is provided. Finally, algorithms to retrieve the bottom reflectance are applied to simulated data and actual remote sensing imagery and their performance is compared. The available methods are not able to completely eliminate the water column effect, but they can minimize its influence. Choosing the best method depends on the marine environment, available input data and desired outcome or scientific application.

Rudorff, ND, Frouin R, Kampel M, Goyens C, Meriaux X, Schieber B, Mitchell BG.  2014.  Ocean-color radiometry across the Southern Atlantic and Southeastern Pacific: Accuracy and remote sensing implications. Remote Sensing of Environment. 149:13-32.   10.1016/j.rse.2014.03.029   AbstractWebsite

Ocean color radiometry (OCR) provides valuable data for biogeochemical oceanography. In situ OCR measurements are used in the development and validation of bio-optical models and vicarious calibration of satellite ocean-color sensors. It is thus crucial to obtain accurate in situ OCR measurements, which is a challenge, especially in regions subjected to adverse environmental conditions and where waters are optically complex. In the present work, the accuracy of in situ OCR is analyzed with data acquired in a wide range of bio-geographic provinces across the Southern Atlantic and Southeastern Pacific during the R/V Melville MV1102 cruise. Varied techniques employed to measure above-water remote sensing reflectance (R-rs) are inter-compared. Measured R is also compared with modeled R-rs in a closure experiment. The impact of R-rs uncertainties on the retrieval of chlorophyll a concentration (Chla) and inherent optical properties (IOPs) is evaluated using operational bio-optical algorithms. The relative percent difference (RPD) between R-rs measured by the various techniques ranged from 12 to 26% for the ocean-color bands (412-555 nm), and 3-12% for the ratios (412-510/555). A merged R-rs obtained by averaging the different types of measurements, INS, is recommended to reduce uncertainties. The coefficient of variation of INS and reflectance ratios was 11-13% and 3-5%, respectively. The RPD between INS and modeled R-rs and the corresponding ratios ranged from 18 to 34% and from 13 to 17%, respectively. Complete closure could not be obtained due to both measurement and modeling uncertainties. The impact of INS uncertainties on retrieved Chla and IOPs was generally smaller than the intrinsic errors of the inversion schemes. The results suggest that even though more accurate ocean-color radiometry is desirable, improving retrieval algorithms is essential to properly describing and furthering our understanding of bio-optical variability in the world's oceans. (C) 2014 Elsevier Inc All rights reserved.

Bastidas-Salamanca, M, Gonzalez-Silvera A, Millan-Nunez R, Santamaria-del-Angel E, Frouin R.  2014.  Bio-optical characteristics of the northern Gulf of California during June 2008. International Journal of Oceanography. :384618(13pp.)-384618(13pp.).   10.1155/2014/384618   AbstractWebsite

Bio-optical variables in the Northern Gulf of California were analyzed using in situ and satellite data obtained during a cruise in June 2008. The study area was divided into three bio-optical regions: Upper Gulf (UG), Northern Gulf (NG), and Great Isles (GI). Each region was characterized according to phytoplankton pigment concentration, phytoplankton and nonpigmented material spectral absorption coefficients, and spectral reflectance. Observed patterns were an indication of the shift in bio-optical conditions from north to south going from turbid and eutrophic waters to mesotrophic ones. Although there was a good agreement between satellite and in situ Chl a (RMSE +or-33%), an overestimation of in situ Chl a was observed. This was partly explained by the presence of nonalgal particles, as well as the influence of desert and continental aerosols, which is generally overcorrected in the standard processing. The UG and NG could be considered as Case 2 waters, but they did exhibit different bio-optical characteristics. This implies that both biological and optical properties should be invoked to better understand water reflectance variability in the study region and its implications for the remote sensing of Chl a and biogeochemical processes.

Frouin, R.  2013.  In-flight Calibration of Satellite Ocean-Colour Sensors. IOCCG Report. ( Frouin R, Ed.).:126., Dartmouth, Canada: IOCCG Abstract
Frouin, R, McPherson J.  2012.  Estimating photosynthetically available radiation at the ocean surface from GOCI data. Ocean Science Journal. 47:313-321.   10.1007/s12601-012-0030-6   AbstractWebsite

A technique is presented to estimate photosynthetically available radiation (PAR) at the ocean surface from Geostationary Ocean Color Imager (GOCI) data. The sensor is adapted to the problem, since it measures at visible wavelengths and does not saturate over clouds, and the hourly data provides adequate temporal sampling to describe diurnal variability of clouds. Instantaneous surface PAR is computed as the difference between the solar irradiance incident at the top of the atmosphere (known) and the solar irradiance reflected back to space (derived from GOCI radiance), taking into account absorption and scattering by the clear atmosphere (modeled). Knowledge of pixel composition is not required. Apart from planetary albedo and sun zenith angle, the model parameters are fixed at their climatological values. The instantaneous PAR estimates at hourly intervals are integrated over time to provide daily values. The technique is applied to GOCI imagery acquired on 5 April 2011, and the GOCI daily PAR estimates are compared with those obtained from MODerate Resolution Imaging Spectrometer (MODIS) data. Agreement is good between the two types of estimates, with a coefficient of determination (r(2)) of 0.778, a bias of 0.23 Em(-2)d(-1) (0.5% with higher GOCI values), and a root-mean-squared difference of 5.00 Em(-2)d(-1) (11.2%). Differences in cloudy conditions are attributed to daily cloudiness changes not captured by the MODIS observations. The comparison statistics indicate that GOCI PAR estimates have acceptable accuracy for regional studies of aquatic photosynthesis.

Montes-Hugo, M, Sweeney C, Doney SC, Ducklow H, Frouin R, Martinson DG, Stammerjohn S, Schofield O.  2010.  Seasonal forcing of summer dissolved inorganic carbon and chlorophyll a on the western shelf of the Antarctic Peninsula. Journal of Geophysical Research-Oceans. 115   10.1029/2009jc005267   AbstractWebsite

The Southern Ocean is a climatically sensitive region that plays an important role in the regional and global modulation of atmospheric CO(2). Based on satellite-derived sea ice data, wind and cloudiness estimates from numerical models (National Centers for Environmental Prediction-National Center for Atmospheric Research reanalysis), and in situ measurements of surface (0-20 m depth) chlorophyll a (Chl(Surf)) and dissolved inorganic carbon (DIC(Surf)) concentration, we show sea ice concentration from June to November and spring wind patterns between 1979 and 2006 had a significant influence on midsummer (January) primary productivity and carbonate chemistry for the Western Shelf of the Antarctic Peninsula (WAP, 64 degrees-68 degrees S, 63.4 degrees-73.3 degrees W). In general, strong (>3.5 m s(-1)) and persistent (> 2 months) northerly winds during the previous spring were associated with relatively high (monthly mean > 2 mg m(-3)) Chl(Surf) and low (monthly mean <2 mmol kg(-1)) salinity-corrected DIC (DIC(Surf)*) during midsummer. The greater Chl(Surf) accumulation and DIC(Surf)* depletion was attributed to an earlier growing season characterized by decreased spring sea ice cover or nearshore accumulation of phytoplankton in association with sea ice. The impact of these wind-driven mechanisms on Chl(Surf) and DIC(Surf)* depended on the extent of sea ice area (SIA) during winter. Winter SIA affected phytoplankton blooms by changing the upper mixed layer depth (UMLD) during the subsequent spring and summer (December-January-February). Midsummer DIC(Surf)* was not related to DIC(Surf)* concentration during the previous summer, suggesting an annual replenishment of surface DIC during fall/winter and a relatively stable pool of deep (> 200 m depth) "winter-like" DIC on the WAP.

Dupouy, C, Neveux J, Ouillon S, Frouin R, Murakami H, Hochard S, Dirberg G.  2010.  Inherent optical properties and satellite retrieval of chlorophyll concentration in the lagoon and open ocean waters of New Caledonia. Marine Pollution Bulletin. 61:503-518.   10.1016/j.marpolbul.2010.06.039   AbstractWebsite

The retrieval of chlorophyll-a concentration from remote sensing reflectance (Rrs) data was tested with the NASA 0C4v4 algorithm on the inner New Caledonian lagoon (Case 2) and adjacent open ocean (Case 1) waters. The input to 0C4v4 was Rrs measured in situ or modeled from water's inherent optical properties (2001-2007). At open ocean stations, backscattering and absorption coefficients were correlated with chlorophyll (R(2) = 0.31-0.51, respectively), in agreement with models for Case 1 waters. Taking spectrofluorometric measurement as reference, the 0C4v4 model leads to an average underestimation of 33% of the chlorophyll concentration. For the lagoon waters, 0C4v4 performed inadequately because the backscattering coefficient, highly correlated with turbidity and suspended matter (R2 = 0.98), was poorly correlated to chlorophyll (R(2) = 0.42). The 0C4v4 performance was better in deep lagoon waters for stations with a TOT index (Tchla x depth/turbidity) higher than 19 mg M-2 NTU-1 (R(2) = 0.974, bias = 10.2%). Global Imager Rrs provided a good estimate of Tchla (R(2) = 0.79, N = 28) in the deeper part of the lagoon. (C) 2010 Elsevier Ltd. All rights reserved.

Dubuisson, P, Frouin R, Dessailly D, Duforet L, Leon JF, Voss K, Antoine D.  2009.  Estimating the altitude of aerosol plumes over the ocean from reflectance ratio measurements in the O(2) A-band. Remote Sensing of Environment. 113:1899-1911.   10.1016/j.rse.2009.04.018   AbstractWebsite

A methodology is proposed to infer the altitude of aerosol plumes over the ocean from reflectance ratio measurements in the O(2) absorption A-band (759 to 770 nm). The reflectance ratio is defined as the ratio of the reflectance in a first spectral band, strongly attenuated by O(2) absorption, and the reflectance in a second spectral band, minimally attenuated. For a given surface reflectance, simple relations are established between the reflectance ratio and the altitude of an aerosol layer, as a function of atmospheric conditions and the geometry of observation. The expected accuracy for various aerosol loadings and models is first quantified using an accurate, high spectral resolution, radiative transfer model that fully accounts for interactions between scattering and absorption. The method is developed for POLDER and MERIS, satellite sensors with adequate spectral characteristics. The simulations show that the method is only accurate over dark surfaces when aerosol optical thickness at 765 nm is relatively large (>0.3). In this case, the expected accuracy is on the order of +/- 0.5 km or +/- 0.2 km for POLDER or MERIS respectively. More accurate estimates are obtained with MERIS, since in this case the spectral reflectance ratio is more sensitive to aerosol altitude. However, a precise spectral calibration is needed for MERIS. The methodology is applied to MERIS and POLDER imagery acquired over marine surfaces. The estimated aerosol altitude is compared with in situ lidar profiles of backscattering coefficient measured during the AOPEX-2004 experiment for MERIS, or obtained with the space-borne lidar CALIOP for POLDER. The retrieved altitudes agree with lidar measurements in a manner consistent with theory. These comparisons demonstrate the potential of the differential absorption methodology for obtaining information on aerosol altitude over dark surfaces. (C) 2009 Elsevier Inc. All rights reserved.

Frouin, R, Pelletier B.  2009.  Consistency of ridge function fields for varying nonparametric regression. Communications in Statistics-Theory and Methods. 38:1272-1283.   10.1080/03610920802395702   AbstractWebsite

A nonparametric regression model proposed in Pelletier and Frouin (2006) as a solution to the geophysical problem of ocean color remote sensing is studied. The model, called ridge function field, combines a regression estimate in the form of a superposition of ridge functions, or which is equivalent to a neural network, with the idea pertaining to varying-coefficients models, where the parameters of a parametric family are allowed to vary with other variables. Under mild assumptions on the underlying distribution of the data, the strong universal consistency of the least-squares ridge function fields estimate is established.

Frouin, R, Pelletier B.  2009.  Function field methodology for estimating spectral marine reflectance from ADEOS-II Global Imager data. Journal of The Remote Sensing Society of Japan. 29:86-95.   10.11440/rssj.29.86   Abstract

A general function field methodology for estimating ocean color variables from space is applied to the retrieval of spectral marine reflectance from Global Imager (GLI) data. The top-of-atmosphere GLI reflectance vectors, after correction for molecular effects, are considered as explanatory variables conditioned by the angular geometry. The inverse problem, therefore, is viewed as a collection of similar inverse problems, continuously indexed by the angular variables. The solution is in the form of a field of nonlinear regression models over the set of permitted values for the angular variables. The selected models, for reasons of approximation theory, are fields of shifted ridge functions. The fields constructed on synthetic GLI data for Case 1 waters are robust to noise, they handle well situations of weakly and strongly absorbing aerosols, and the retrievals are accurate in both oligotrophic and productive waters. In the presence of 1% noise, the RMS error is 0.0006 (4.2%) at 380 nm, 0.0003 (2.8%) at 460nm, and 0.0001 (1.5%) at 545nm, i.e., well within the acceptable limits for quantitative biology applications. The theoretical results, and the possible extensions, show the potential of the function field methodology for operational estimation of marine reflectance from GLI data, even in the near ultraviolet.

Duforet, L, Frouin R, Dubuisson P.  2007.  Importance and estimation of aerosol vertical structure in satellite ocean-color remote sensing. Applied Optics. 46:1107-1119.   10.1364/ao.46.001107   AbstractWebsite

The vertical distribution of absorbing aerosols affects the reflectance of the ocean-atmosphere system. The effect, due to the coupling between molecular scattering and aerosol absorption, is important in the visible, especially in the blue, where molecular scattering is effective, and becomes negligible in the near infrared. It increases with increasing Sun and view zenith angles and aerosol optical thickness and with decreasing scattering albedo but is practically independent of wind speed. Relative differences between the top of the atmosphere reflectance simulated with distinct vertical distributions may reach approximately 10% or even 20%, depending on aerosol absorption. In atmospheric correction algorithms, the differences are directly translated into errors on the retrieved water reflectance. These errors may reach values well above the 5 x 10(-4) requirement in the blue, even for small aerosol optical thickness, preventing accurate retrieval of chlorophyll-a [Chl-a] concentration. Estimating aerosol scale height or altitude from measurements in the oxygen A band, possible with the polarization and directionality of the Earth's reflectance instrument and medium resolution imaging spectrometer, is expected to improve significantly the accuracy of the water reflectance retrievals and yield acceptable [Chl-a] concentration estimates in the presence of absorbing aerosols.

Frouin, R, Murakami H.  2007.  Estimating photosynthetically available radiation at the ocean surface from ADEOS-II Global Imager data. Journal of Oceanography. 63:493-503.   10.1007/s10872-007-0044-3   AbstractWebsite

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, i.e., the planetary atmosphere is modeled as a clear atmosphere positioned above a cloud layer. PAR is computed as the difference between the incident 400-700 urn solar flux at the top of the atmosphere (known) and the solar flux reflected back to space by the atmosphere and surface (derived from GLI radiance), taking atmospheric absorption into account. 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 Earth Radiation Budget Satellite (ERBS) data. The algorithm results are verified against other satellite estimates of PAR, the National Centers for Environmental Prediction (NCEP) reanalysis product, and in-situ measurements from fixed buoys. Agreement is generally good between GLI and Sea-viewing Wide Field-of-view Sensor (SeaWiFS) estimates, with root-mean-squared (rms) differences of 7.9 (22%), 4.6 (13%), and 2.7 (8%) Einstein/m(2)/day on daily, weekly, and monthly time scales, and a bias of only 0.8-0.9 (about 2%) Einstein/m(2)/day. The rms differences between GLI and Visible and Infrared Spin Scan Radiometer (VISSR) estimates and between GLI and NCEP estimates are smaller and larger, respectively, on monthly time scales, i.e., 3.0 (7%) and 5.0 (14%) Einstein/m(2)/day, and biases are 1.1 (2%) and -0.2 (-1%) Einstein/m(2)/day. The comparison with buoy data also shows good agreement, with rms inaccuracies of 10.2 (23%), 6.3 (14%), and 4.5 (10%) Einstein/m(2)/day on daily, weekly, and monthly time scales, and slightly higher GLI values by about 1.0 (2%) Einstein/m(2)/day. The good statistical performance makes the algorithm suitable for large-scale studies of aquatic photosynthesis.

Fukushima, H, Toratani M, Murakami H, Deschamps PY, Frouin R, Tanaka A.  2007.  Evaluation of ADEOS-II GLI ocean color atmospheric correction using SIMBADA handheld radiometer data. Journal of Oceanography. 63:533-543.   10.1007/s10872-007-0048-z   AbstractWebsite

The performance of the "version 2" Global Imager (GLI) standard atmospheric correction algorithm, which includes empirical absorptive aerosol correction and sun glint correction, was evaluated using data collected with handheld above-water SIMBADA radiometers during 23 cruises of opportunity (research vessels, merchant ships), mostly in the North Atlantic and European seas. A number of 100 match-up data sets of GLI-derived and SIMBADA-measured normalized water-leaving radiance (nL(w)) and aerosol optical thickness (AOT) were sorted out, using objective selection criteria, and analyzed. The Root-Mean-Square (RMS) difference between GLI and SIMBADA nL(w) was about 0.32 mu W/cm(2)/nm/sr for the 412 nm band, showing improvement by 30% in RMS difference with respect to the conventional "version V GLI atmospheric correction algorithm, and the mean difference (or bias) was reduced significantly. For AOT, the RMS difference was 0.1 between GLI estimates and SIMBADA measurements and the bias was small (a few 0.01), but the Angstrom exponent was systematically underestimated, by 0.4 on average, suggesting a potential GLI calibration offset in the near infrared. The nL(w) differences were not correlated to AOT, although performance was best in very clear conditions (AOT less than 0.05 in the 865 nm band). Despite the relatively large scatter between estimated and measured nL(w) the derived chlorophyll-a concentration estimates, applying the same ratio algorithm (GLI OC4V4) to GLI and SIMBADA, were consistent and highly correlated in the range of 0.05-2 mu g/l. The large variability in chlorophyll-a concentration estimate for clear clean water areas (e.g. with the concentration range lower than about 0.05 mu g/l) turns out to be due to the nature of the "band ratio" based in-water algorithm.

Frouin, R, Pelletier B.  2007.  Fields of non-linear regression models for atmospheric correction of satellite ocean-color imagery. Remote Sensing of Environment. 111:450-465.   10.1016/j.rse.2007.04.005   AbstractWebsite

Remote sensing of ocean color from space, a problem that consists of retrieving spectral marine reflectance from spectral top-of-atmosphere 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 marine reflectance. Since the spectral components of the field take values in the same variable vector space, the retrievals in individual spectral bands are not independent, i.e., the solution is not just a juxtaposition of independent models for each spectral band. A scheme based on ridge functions is developed to approximate this solution to an arbitrary accuracy, and is applied to the retrieval of marine reflectance in Case 1 waters, for which optical properties are only governed by biogenic content. The statistical models are evaluated on synthetic data as well as actual data originating from the SeaWiFS instrument, taking into account noise in the data. Theoretical performance is good in terms of accuracy, robustness, and generalization capabilities, suggesting that the function field methodology might improve atmospheric correction in the presence of absorbing aerosols and provide more accurate estimates of marine reflectance in productive waters. When applied to SeaWiFS imagery acquired off California, the function field methodology gives generally higher estimates of marine reflectance than the standard SeaDAS algorithm, but the values are more realistic. (c) 2007 Elsevier Inc. All rights reserved.

Frouin, R, Deschamps PY, Gross-Colzy L, Murakami H, Nakajima TY.  2006.  Retrieval of chlorophyll-a concentration via linear combination of ADEOS-II Global Imager data. Journal of Oceanography. 62:331-337.   10.1007/s10872-006-0058-2   AbstractWebsite

Top-of-atmosphere reflectance measured above the ocean in the visible and near infrared, after correction for molecular scattering, may be linearly combined to retrieve surface chlorophyll-a abundance directly, without explicit correction for aerosol scattering and absorption. The coefficients of the linear combination minimize the perturbing effects, which are modeled by a polynomial, and they do not depend on geometry. The technique has been developed for Global Imager (GLI) spectral bands centered at 443, 565, 667, and 866 nm, but it is applicable to other sets of spectral bands. Theoretical performance is evaluated from radiation-transfer simulations for a wide range of geophysical and angular conditions. Using a polynomial with exponents of -2, -1, and 0 to determine the coefficients, the residual influence of the atmosphere on the linear combination is within +/- 0.001 in most cases, allowing chlorophyll-a abundance to be retrieved with a root-mean-squared (RMS) error of 8.4% in the range 0.03-3 mgm(-3). Application of the method to simulated GLI imagery shows that estimated and actual chlorophyll-a abundance are in agreement, with an average RMS difference of 32.1% and an average bias of -2.2% (slightly lower estimated values). The advantage of the method resides in its simplicity, flexibility, and rapidity of execution. Knowledge of aerosol amount and type is avoided. There is no need for look-up tables of aerosol optical properties. Accuracy is adequate, but depends on the polynomial representation of the perturbing effects and on the bio-optical model selected to relate the linear combination to chlorophyll-a abundance. The sensitivity of the linear combination to chlorophyll-a abundance can be optimized, and the method can be extended to the retrieval of other bio-optical variables.

Murakami, H, Sasaoka K, Hosoda K, Fukushima H, Toratani M, Frouin R, Mitchell BG, Kahru M, Deschamps PY, Clark D, Flora S, Kishino M, Saitoh S, Asanuma I, Tanaka A, Sasaki H, Yokouchi K, Kiyomoto Y, Saito H, Dupouy C, Siripong A, Matsumura S, Ishizaka J.  2006.  Validation of ADEOS-II GLI ocean color products using in-situ observations. Journal of Oceanography. 62:373-393.   10.1007/s10872-006-0062-6   AbstractWebsite

The Global Imager (GLI) aboard the Advanced Earth Observing Satellite-II (ADEOS-II) made global observations from 2 April 2003 to 24 October 2003. In cooperation with several institutes and scientists, we obtained quality controlled match-ups between GLI products and in-situ data, 116 for chlorophyll-a concentration (CHLA), 249 for normalized water-leaving radiance (nLw) at 443 nm, and 201 for aerosol optical thickness at 865 nm (Tau_865) and Angstrom exponent between 520 and 865 nm (Angstrom). We evaluated the GLI ocean color products and investigated the causes of errors using the match-ups. The median absolute percentage differences (MedPD) between GLI and in-situ data were 14.1-35.7% for nLws at 380-565 nm 52.5-74.8% nLws at 625-680 nm, 47.6% for Tau_865, 46.2% for Angstrom, and 46.6% for CHLA, values that are comparable to the ocean-color products of other sensors. We found that some errors in GLI products are correlated with observational conditions; nLw values were underestimated when nLw at 680 nm was high, CHLA was underestimated in absorptive aerosol conditions, and Tau_865 was overestimated in sunglint regions. The error correlations indicate that we need to improve the retrievals of the optical properties of absorptive aerosols and seawater and sea surface reflection for further applications, including coastal monitoring and the combined use of products from multiple sensors.

Pelletier, B, Frouin R.  2006.  Remote sensing of phytoplankton chlorophyll-a concentration by use of ridge function fields. Applied Optics. 45:784-798.   10.1364/ao.45.000784   AbstractWebsite

A methodology is presented for retrieving phytoplankton chlorophyll-alpha concentration from space. The data to be inverted, namely, vectors of top-of-atmosphere reflectance in the solar spectrum, are treated as explanatory variables conditioned by angular geometry. This approach leads to a continuum of inverse problems, i.e., a collection of similar inverse problems continuously indexed by the angular variables. The resolution of the continuum of inverse problems is studied from the least-squares viewpoint and yields a solution expressed as a function field over the set of permitted values for the angular variables, i.e., a map defined on that set and valued in a subspace of a function space. The function fields of interest, for reasons of approximation theory, are those valued in nested sequences of subspaces, such as ridge function approximation spaces, the union of which is dense. Ridge function fields constructed on synthetic yet realistic data for case I waters handle well situations of both weakly and strongly absorbing aerosols, and they are robust to noise, showing improvement in accuracy compared with classic inversion techniques. The methodology is applied to actual imagery from the Sea-Viewing Wide Field-of-View Sensor (SeaW-iFS); noise in the data are taken into account. The chlorophyll-alpha concentration obtained with the function field methodology differs from that obtained by use of the standard SeaWiFS algorithm by 15.7% on average. The results empirically validate the underlying hypothesis that the inversion is solved in a least-squares sense. They also show that large levels of noise can be managed if the noise distribution is known or estimated. (c) 2006 Optical Society of America.