Charles R. McClain

Ocean color chlorophyll algorithms for SeaWiFS

A large data set containing coincident in situ chlorophyll and remote sensing reflectance measurements was used to evaluate the accuracy, precision, and suitability of a wide variety of ocean color chlorophyll algorithms for use by SeaWiFS (Sea-viewing Wide Field-of-view Sensor). The radiance-chlorophyll data were assembled from various sources during the SeaWiFS Bio-optical Algorithm Mini-Workshop (SeaBAM) and is composed of 919 stations encompassing chlorophyll concentrations between 0.019 and 32.79 μg L−1. Most of the observations are from Case I nonpolar waters, and ∼20 observations are from more turbid coastal waters. A variety of statistical and graphical criteria were used to evaluate the performances of 2 semianalytic and 15 empirical chlorophyll/pigment algorithms subjected to the SeaBAM data. The empirical algorithms generally performed better than the semianalytic. Cubic polynomial formulations were generally superior to other kinds of equations. Empirical algorithms with increasing complexity (number of coefficients and wavebands), were calibrated to the SeaBAM data, and evaluated to illustrate the relative merits of different formulations. The ocean chlorophyll 2 algorithm (OC2), a modified cubic polynomial (MCP) function which uses Rrs490/Rrs555, well simulates the sigmoidal pattern evident between log-transformed radiance ratios and chlorophyll, and has been chosen as the at-launch SeaWiFS operational chlorophyll a algorithm. Improved performance was obtained using the ocean chlorophyll 4 algorithm (OC4), a four-band (443, 490, 510, 555 nm), maximum band ratio formulation. This maximum band ratio (MBR) is a new approach in empirical ocean color algorithms and has the potential advantage of maintaining the highest possible satellite sensor signal: noise ratio over a 3-orders-of-magnitude range in chlorophyll concentration.

Asian dust events of April 1998

On April 15 and 19, 1998, two intense dust storms were generated over the Gobi desert by springtime low-pressure systems descending from the northwest. The windblown dust was detected and its evolution followed by its yellow color on SeaWiFS satellite images, routine surface-based monitoring, and through serendipitous observations. The April 15 dust cloud was recirculating, and it was removed by a precipitating weather system over east Asia. The April 19 dust cloud crossed the Pacific Ocean in 5 days, subsided to the surface along the mountain ranges between British Columbia and California, and impacted severely the optical and the concentration environments of the region. In east Asia the dust clouds increased the albedo over the cloudless ocean and land by up to 10–20%, but it reduced the near-UV cloud reflectance, causing a yellow coloration of all surfaces. The yellow colored backscattering by the dust eludes a plausible explanation using simple Mie theory with constant refractive index. Over the West Coast the dust layer has increased the spectrally uniform optical depth to about 0.4, reduced the direct solar radiation by 30–40%, doubled the diffuse radiation, and caused a whitish discoloration of the blue sky. On April 29 the average excess surface-level dust aerosol concentration over the valleys of the West Coast was about 20–50 μg/m3 with local peaks >100 μg/m3. The dust mass mean diameter was 2–3 μm, and the dust chemical fingerprints were evident throughout the West Coast and extended to Minnesota. The April 1998 dust event has impacted the surface aerosol concentration 2–4 times more than any other dust event since 1988. The dust events were observed and interpreted by an ad hoc international web-based virtual community. It would be useful to set up a community-supported web-based infrastructure to monitor the global aerosol pattern for such extreme aerosol events, to alert and to inform the interested communities, and to facilitate collaborative analysis for improved air quality and disaster management.

Distributions of phytoplankton blooms in the southern ocean.

A regional pigment retrieval algorithm for the Nimbus-7 Coastal Zone Color Scanner (CZCS) has been tested for the Southern Ocean. The pigment concentrations estimated with this algorithm agree to within 5 percent with in situ values and are more than twice as high as those previously reported. The CZCS data also revealed an asymmetric distribution of enhanced pigments in the waters surrounding Antarctica; in contrast, most surface geophysical properties are symmetrically distributed. The asymmetry is coherent with circumpolar current patterns and the availability of silicic acid in surface waters. Intense blooms (>1 milligram of pigment per cubic meter) that occur downcurrent from continental masses result from dissolved trace elements such as iron derived from shelf sediments and glacial melt.

Coastal zone color scanner pigment concentrations in the southern ocean and relationships to geophysical surface features

The spatial and seasonal distributions of phytoplankton pigment concentration over the entire southern ocean have been studied for the first time using the coastal zone color scanner historical data set (from October 1978 through June 1986). Enhanced pigment concentrations are observed between 35°S and 55°S throughout the year, with such enhanced regions being more confined to the south in the austral summer and extending further north in the winter. North and south of the polar front, phytoplankton blooms (>1 mg/m3) are not uniformly distributed around the circumpolar region. Instead, blooms appear to be located in regions of ice retreat (or high melt areas) such as the Scotia Sea and the Ross Sea, in relatively shallow areas (e.g., the Patagonian and the New Zealand shelves), in some regions of Ekman upwelling like the Tasman Sea, and near areas of high eddy kinetic energy such as the Agulhas retroflection. Among all features examined by regression analysis, bathymetry appears to be the one most consistently correlated with pigments (correlation coefficient being about −0.3 for the entire region). The cause of negative correlation with bathymetry is unknown but is consistent with the observed abundance of iron in shallow areas in the Antarctic region. It is also consistent with resuspension of phytoplankton cells by wind-induced mixing, especially in shallow waters. On the other hand, in the deep ocean (especially at latitudes 30%) in average pigment concentration over the entire region during different seasons indicates possible influence of time dependent parameters.

The calibration and validation of SeaWiFS data

Abstract The Sea-viewing Wide Field-of-view Sensor (SeaWiFS) is the successor ocean color imaging system to the Nimbus-7 Coastal Zone Color Scanner (CZCS). The SeaWiFS calibration and validation effort includes spacecraft, atmospheric, sea surface, subsurface (or in situ), plus laboratory and data analysis components which require pre- and postlaunch activities. The most important goals of this effort are to produce water-leaving radiances with an uncertainty of 5% in clear-water regions and chlorophyll a concentrations within ±35% over the range of 0.05–50 mg m−3. The first objective requires field instruments with a calibration and measurement capability on the order of 1%; because these challenging in situ measurements will be acquired from a variety of field instruments over the five-year mission interval, a measurement assurance program is required. This program consists of several activities: an accurate prelaunch characterization and calibration of the SeaWiFS instrument; a Marine Optical Buoy (MOBY) rotation in clear water to provide a water-leaving radiance time series for postlaunch vicarious calibration; the SeaWiFS Bio-Optical Archive and Storage System (SeaBASS) to hold the relevant data; clearly defined SeaWiFS Ocean Optics Protocols (SOOP) for established data collection methodologies; annual SeaWiFS Intercalibration Round-Robin Experiments (SIRREXs) for intercomparing field and calibration equipment, and training scientific personnel; direct comparison to a national standard laboratory using the SeaWiFS Transfer Radiometer (SXR); a portable field source, called the SeaWiFS Quality Monitor (SQM), for monitoring the temporal stability of the calibration of field instruments; a highly accurate atmospheric correction algorithm designed for the SeaWiFS instrument response functions; bio-optical algorithms that encompass a broad range of bio-optical provinces; and satellite data processing, quality control, and analysis procedures for monitoring the postlaunch performance of the sensor and the validity of the derived products. The culmination of many of these activities is the deployment of the instruments and methodologies on Atlantic Meridional Transect (AMT) cruises between England and the Falkland Islands, a 13 000 km voyage spanning more than 100° of latitude, with a calibration and measurement accuracy that is on the order of 1%. The AMT Program is the primary product validation activity supported by the SeaWiFS Project. The AMT cruises also serve as a testbed for new technology development and have demonstrated that high quality bio-optical data can be routinely provided to the Project in near-real time.

Pigment distribution in the Caribbean sea: Observations from space

Abstract The Caribbean is a semi-enclosed tropical sea which is generally considered oligotrophic, but that is influenced by nearly 20% of the annual discharge of the worlds rivers (Amazon and Orinoco Rivers) and by seasonal upwelling along the southern margin. To investigate the role of these nutrient sources on the productivity of the region, we mapped the distribution of pigments in the eastern Caribbean (east of 80° W) using a series of Coastal Zone Color Scanner (CZCS) images collected between November 1978 and December 1982. Five additional images were examined for the period 1983–1986. The images revealed a seasonal cycle in the spatial structure of near-surface pigment. During January-May, there were high pigment concentrations (> 0.5 mg m−3) along the continental margin (south of 14°N), where upwelling occurred. Very little pigment ( 3 × 105 km2). This plume reached Puerto Rico around September-October and drifted westward, slowly losing its color signature. We estimate that the discharge of the Orinoco contributes 2–12% of the daily nitrogen requirements of the phytoplankton growing in the river plume, and leads to the fixation of 7–29 × 105 tons of carbon per year. The rest of the nitrogen demand appears to be met by nitrogen cycling. The large-scale (> 100 km) pigment distribution patterns in the Caribbean Sea seem to be controlled by wind stress, flux of water through the basin, and river discharge. Westward advection of Atlantic water probably dominates the flow during the first half of the year, restricting the dispersal of blooms to the southern half of the Caribbean while flushing the central and northern portions. As the influx of Atlantic water decreases in the second half of the year, local Ekman transport driven by the trade winds becomes dominant and surface waters drift northwestward throughout the basin. The seasonal sequence of changes in pigment distribution patterns was consistent from year to year except in 1980 and 1984, when wind conditions in the Caribbean were anomalous. Close scrutiny of the 4 years of CZCS images did not reveal any evidence of large-scale (> 300 km) eastward-flowing currents in the central Caribbean. This supports the view that previous observations of countercurrents were based on partial sampling of eddies of 100–250 km diameter.

Ocean color variability of the tropical Indo‐Pacific basin observed by SeaWiFS during 1997–1998

High-quality ocean color data (chlorophyll) provided by the Sea-viewing Wide Field of view Sensor (SeaWiFS) satellite were analyzed for the first complete year of coverage (October 1997 to September 1998) in the tropical Indo-Pacific basin. This period coincides with the peak of one of the strongest El Nino events during December 1997 and the La Nina of 1998 that appeared dramatically in less than a month as a sea surface temperature (SST) change of over 6°C in the central equatorial Pacific during June 1998. The tropical Indian Ocean also underwent a highly anomalous series of events with negative SST anomalies (SSTA) of over 3°C in the eastern equatorial and coastal regions during October-December 1997 and warm SSTA in the west that peaked at over 2°C during February 1998. The ocean color variability is interpreted using other satellite data such as sea level from TOPEX/Poseidon and also in terms of the dynamics and thermodynamics of the region from simulations with an ocean general circulation model. The El Nino-related reductions in equatorial production and the off-equatorial increase in biological activity, and their basin scale evolution is clearly seen for the first time. Persistent northerly wind anomalies resulted in a northward shift of the equatorial divergence and the upwelling Kelvin wave which signalled the end of the 1997–1998 El Nino. The anomalous surface chlorophyll associated with this Kelvin wave was also clearly shifted north of the equator by nearly 300 km and appeared more than a month before the negative sea level anomalies seen by TOPEX/Poseidon. On the equator near 165°E, the disappearance of the barrier layer appeared to coincide with a localized bloom that occurred in response to the easterly wind bursts over the western Pacific that lasted from December 1997 through the boreal summer. The ecosystem response to the cold La Nina conditions is clearly seen as elevated chlorophyll during the boreal summer of 1998 in the equatorial Pacific cold tongue region. In the Indian Ocean, an anomalous phytoplankton bloom was observed by SeaWiFS during October-December 1997 coincident with the anomalous upwelling in the eastern equatorial region and off the coast of Sumatra. A stronger than normal northeast monsoon is seen as higher than climatological values of surface chlorophyll. The open ocean Ekman pumping and the shoaling of the thermocline near 60°E and 10°S and the eastward extension of mixed layer entrainment in the same latitude band is seen as a region of higher biological activity during the boreal summer.

Sensor-independent approach to the vicarious calibration of satellite ocean color radiometry

The retrieval of ocean color radiometry from space-based sensors requires on-orbit vicarious calibration to achieve the level of accuracy desired for quantitative oceanographic applications. The approach developed by the NASA Ocean Biology Processing Group (OBPG) adjusts the integrated instrument and atmospheric correction system to retrieve normalized water-leaving radiances that are in agreement with ground truth measurements. The method is independent of the satellite sensor or the source of the ground truth data, but it is specific to the atmospheric correction algorithm. The OBPG vicarious calibration approach is described in detail, and results are presented for the operational calibration of SeaWiFS using data from the Marine Optical Buoy (MOBY) and observations of clear-water sites in the South Pacific and southern Indian Ocean. It is shown that the vicarious calibration allows SeaWiFS to reproduce the MOBY radiances and achieve good agreement with radiometric and chlorophyll a measurements from independent in situ sources. We also find that the derived vicarious gains show no significant temporal or geometric dependencies, and that the mission-average calibration reaches stability after approximately 20-40 high-quality calibration samples. Finally, we demonstrate that the performance of the vicariously calibrated retrieval system is relatively insensitive to the assumptions inherent in our approach.

Annual cycles of phytoplankton chlorophyll concentrations in the global ocean: A satellite view

Conceptual and mathematical models show that annual cycles of phytoplankton biomass are different within different regions of the ocean. The purpose of this manuscript is to use coastal zone color scanner chlorophyll imagery (CZCS-Ch1) to determine annual cycles in phytoplankton chlorophyll (biomass) averaged over very large areas of the global ocean. A possible result is that large-scale averaging of CZCS-Ch1 will yield no interpretable signals because of spatial variability in annual cycles at scales much smaller than our averaging scale. Alternatively, if our analyses show regular and persistent global patterns, then our results will jprovide a basin-scale overview of phytoplankton biomass seasonally for comparison with model results or with other large-scale oceanographic measurements. Our results show that monthly mean CZCS-Ch1 imagery (and using in situ concentrations for winter at latitudes poleward of 40 deg) resolves important differences in annual phytoplankton chlorophyll cycles for different ocean basins and latitude belts. As predicted by simple models of plankton dynamics, our results show: (1) global subtropical waters have circa 2X higher CZCS-Ch1 concentrations in winter than in summer and (2) subpolar waters in the northern hemisphere (NH) have mean monthly CZCS-Ch1 concentrations during May and June that are manyfold higher than in winter, particularly in the North Atlantic. Our results also show: (1) Northern Indian Ocean is the major subtropical anomaly, (2) subpolar waters in the SH do not show differences between spring maxima and winter minima as large as those in the subpolar NH and (3) larger ocean area in the SH is compensated by higher mean annual CZCS-Ch1 concentrations in the NH, so that annual hemispherical integrals (mean annual concentrations multiplied by ocean areas) are very similar. The simple patterns we report imply that mean annual cycles in phytoplankton biomass averaged over very large areas of the global ocean are largely explainable by very simple mathematical models such as those presented several decades ago by Cushing, Riley, Steele, and others.

Effects of Penetrative Radiation on the Upper Tropical Ocean Circulation

Abstract The effects of penetrative radiation on the upper tropical ocean circulation have been investigated with an ocean general circulation model (OGCM) with attenuation depths derived from remotely sensed ocean color data. The OGCM is a reduced gravity, primitive equation, sigma coordinate model coupled to an advective atmospheric mixed layer model. These simulations use a single exponential profile for radiation attenuation in the water column, which is quite accurate for OGCMs with fairly coarse vertical resolution. The control runs use an attenuation depth of 17 m while the simulations use spatially variable attenuation depths. When a variable depth oceanic mixed layer is explicitly represented with interactive surface heat fluxes, the results can be counterintuitive. In the eastern equatorial Pacific, a tropical ocean region with one of the strongest biological activity, the realistic attenuation depths result in increased loss of radiation to the subsurface, but result in increased sea surface te...