Coastal Zone Color Scanner Images Research Articles

위성원격탐사에 의한 동한난류 주변 해역의 색소농도 공간적 분포 -1980년 5월 관측을 중심으로 -

동한난류 주변 해역의 해색전선과 수온전선의 공간적 분포 특성을 CZCS와 AVHRR 위성자료를 이용하여 살펴보았다. 동한난류 주류의 난수 streamer 주변 해역의 색소농도와 SST의 공간적 분포를 살펴본 결과, 식물플랑크톤의 흡수보다는 황색물질이나 용존유기물질과 같은 부유물질에 의해 높은 색소농도를 나타낸 연안냉수해역, 연안의 부유물질에 의한 영향과 식물플랑크톤에 의한 영향이 혼합된 특성이 나타난 색소농도가 높은 수온전선의 냉수해역, 외해수의 특성과 유사한 순수한 식물플랑크톤의 영향에 의해 색소농도가 높은 난수중첩해역, 식물플랑크톤의 홉수에 의해 지배를 받고 있으나 색소농도가 적은 난수해역, 식물플랑크톤의 흡수에 의한 영향으로 색소농도가 높은 외해해역의 5개 범주로 구분할 수 있었다. 동해안을 따른 연안냉수해역의 높은 색소농도의 확장은 동한난류의 흐름에 크게 영향을 받고 있으며, 동한난류의 난수 내부에 있는 낮은 색소농도와 분리된다. 또한, 동한난류해역 주변의 색소 농도와 SST 사이의 관계는 전체적으로, 색소농도가 높은 곳은 수온이 낮은 곳에 나타나고 있으며, 색소농도가 낮은 곳은 수온이 높은 곳에서 나타났다. 색소농도가 가장 높게 나타난 것은 수온전선해역 냉수쪽에서 가장 높은 550nm 반사와 비교적 높은 443nm 흡수와 일치하는 일반적인 연안수와 같이 식물플랑크톤 단독으로 나타나는 것이 아니라 부유물질과 동시에 나타났다. 그리고, 또 다른 색소농도의 높은 농도가 외해수에서 나타난 것과 유사하게 난수 streamer 내부에서 나타났으며, 이것은 연안의 부유물질의 영향은 거의 받지 않은 식물플랑크톤에 의해 색소농도가 높은 것을 알았다. 동한난류 해역과 같이 색소농도가 높은 해역에서 SST와 색소농도 영상만으로 복잡한 물리 생물학적인 현상에 기인한 모든 현상을 이해하기는 어렵지만, 광범위한 해역에서 이러한 현상을 이해하는데 위성자료는 효과적인 수단으로 판단된다. 또한, 현재 운용중인 NASA의 Seastar에 탑재된 SeaWiFS (Sea-viewing Wide Field-of·view Sensor)에 의한 보다 정확한 식물플랑크톤 색소농도 자료는 현장에서 측정된 관측자료와 더불어 동한난류 해역의 물리<TEX>$\cdot$</TEX>생물학적인 시공간변동을 이해하는데 많은 도움을 줄 것이라 사료된다. Spatial distribution of Phytoplankton Pigment Concentration (PPC) and Sea Surface Temperature (SST) around the East Korean Warm Current (EKWC) was described, using both Coastal Zone Color Scanner (CZCS) images and Advanced Very High Resolution Radiometer (AVHRR) images in May, 1980. Water mass in this region can be classified into five categories in the horizontal profile of PPC and SST, nLw (normalized water-leaving radiance) images: (1) coastal cold water region associated with concentrations of dissolved organic material or yellow colored substances and suspended sediments, (2) cold water region of thermal frontal occurred by a combination of phytoplankton absorption and suspended materials, (3) warm water overlay region by the phytoplankton absorption than the suspended materials; (4) warm water region occurred by the low phytoplankton absorption, and (5) offshore region occurred by the high phytoplankton absorption. In particular, the highest PPC (>2.0 mg/m^3) area appeared in the CZCS and AVHRR images with a band shaped distribution of the thermal front and ocean color front region, which is located the coastal cold waters alonB western thermal front of the warm streamer of the EKWC. In this region, the highest PPC occurred by a combination of the high absorption of the phytoplankton (443 nm) and highest reflectance of suspended materials (550 nm). Another high PPC (<TEX>$\simeq$</TEX><TEX>$6\;mg/m^3$</TEX>) appeared in the warm water overlay region inside warm streamer. High phytoplankton pigment concentration of this region was corresponding to the short wavelength of 443 nm, which represented phytoplankton absorption of the CZCS image.

Characteristics of wind‐generated rings in the eastern tropical Pacific Ocean

Eddies are generated in the eastern tropical Pacific (3°S–23°N, 75°–105°W) by winds blowing through Central American mountain passes from the Atlantic. We used Coastal Zone Color Scanner (CZCS) and advanced very high resolution radiometer (AVHRR) satellite imagery complemented with monthly in situ sea surface temperature and wind series from the Comprehensive Ocean‐Atmosphere Data Set (COADS) to study these eddies and their effect on pigment concentrations in the region. Pigment values in the Gulf of Tehuantepec generally reach higher values in November–March before those in the Gulf of Papagayo. The eddies generated in the Gulf of Tehuantepec are associated with passages of cold fronts across the Gulf of Mexico from the north, while the eddies off Papagayo and Panamá are associated with increases in trade wind intensity. CZCS images showed larger numbers of eddies per season than have been previously reported on the basis of in situ and AVHRR observations or numerical simulations. We counted 13 eddies in 1979–1980, 8 in 1984–1985, and 6 in 1985–1986. The eddies transfer both energy and biological constituents from the continental margin to the offshore tropical Pacific. The eddies frequently moved distances in excess of 1500 km from their point of origin. Both anticyclonic and cyclonic eddies are generated, but in general, there are more anticyclones. Anticyclonic eddies generally moved to the southwest. Some cyclonic eddies moved to the south and southeast along the Central American coast and appeared to be trapped by the cyclonic Costa Rica thermal dome. Eddies traveled at speeds varying between 9 and 21 cm s−1 and had diameters of 100–500 km. Phytoplankton concentrations associated with the eddies varied from ∼2 to &gt;10 mg m−3 within ∼70 km of the coast to ∼1 mg m−3 up to 600 km of the coast. Between late April and October, fewer eddies were observed, and phytoplankton concentrations were lower (&lt;0.25 mg m−3) and more uniform over the region.

Temporal and Spatial Variability of Phytoplankton Pigment Concentrations in the Japan Sea Derived from CZCS Images

Temporal and spatial variability of phytoplankton pigment concentrations in the Japan Sea are described, using monthly mean composite images of the Coastal Zone Color Scanner (CZCS). In order to describe the seasonal changes of pigment concentration from the results of the empirical orthogonal function (EOF) analysis, we selected four areas in the south Japan Sea. The pigment concentrations in these areas show remarkable seasonal variations. Two annual blooms appear in spring and fall. The spring bloom starts in the Japan Sea in February and March, when critical depth (CRD) becomes equal to mixed layer depth (MLD). The spring bloom in the southern areas (April) occurs one month in advance of that in the northern areas (May). This indicates that the pigment concentrations in the southern areas may increase rapidly in comparison with the northern areas since the water temperature increases faster in spring in the southern than in the northern areas. The fall bloom appears first in the southwest region, then in the southeast and northeast regions, finally appearing in the northwest region. Fall bloom appears in November and December when MLD becomes equal to CRD. The fall bloom can be explained by deepening of MLD in the Japan Sea. The pigment concentrations in winter are higher than those in summer. The low pigment concentrations dominate in summer.

The onset of the Spring Bloom in the MEDOC area: mesoscale spatial variability

In the northwestern Mediterranean Sea, Coastal Zone Color Scanner images suggest that the eddies that participate in the restratification following deep convection interact with the spring phytoplankton bloom. The mechanisms for this interaction are studied using a biogeochemical model embedded in an eddy-resolving primitive equation ocean model. The model is initialized with a patch of dense water surrounded by a stratified ocean, which is characteristic of the winter situation. The atmospheric forcing is artificially held constant, in order to focus solely on the mesoscale variability. After a few days, meanders develop at the periphery of the patch, inducing its sinking and spreading. Mesoscale upward motions are responsible for the shoaling of the mixing layer in the trough of the meanders. As sunlight is the main factor regulating primary production at this time of year, this shoaling increases the mean exposure time of the phytoplankton cells and thus enhances productivity. Consequently, the majority of phytoplankton production is obtained at the edge of the patch, in agreement with in situ data. Through advection, phytoplankton is then subducted from these sources towards the crest of the meanders. Our results suggest that this mesoscale transport is responsible for a decorrelation between phytoplankton biomass and primary production.

Joint analysis of temperature and ocean colour satellite images for mesoscale activities in the Gulf of Biscay

In the framework of the Ocean Colour European Archive Network (OCEAN) programme of the European Commission and the European Space Agency, a joint team composed of ACRI, GM-Images and Ecole des Mines de Paris proposed to analyse jointly Coastal Zone Color Scanner (CZCS) (chlorophyll) and Advanced Very High Resolution Radiometer (AVHRR) (temperature) data in order to study mesoscale dynamics features in the Gulf of Biscay. A large archive of AVHRR data, between 1978-1990, was analysed with the support of EPSHOM (Etablissement Principal du Service Hydrographique et Oceanographique de la Marine), together with about one hundred pairs of AVHRR and CZCS images. The method used to process AVHRR and CZCS data is presented in this paper. Interesting oceanographic results have been obtained. Mesoscale features such as eddies and plumes have been observed and detailed. Thestructures have been identified and classified according to an accurate typology. The appearance and evolution of mushroom-like structures have been noted, which may promote a high efficiency in the mixing of water masses. The benefit of the fusion of images is discussed. Although the locations of structures observed in both image types differ perceptibly, the increased number of observations leads to better mapping and monitoring of the features, both in space and time.

Enhanced chlorophyll at the shelfbreak of the Mid‐Atlantic Bight and Georges Bank during the spring transition

In 8 yr (1979–1986) of Coastal Zone Color Scanner (CZCS) imagery, we find annual enhancement of chlorophyll at the shelfbreak of the Mid‐Atlantic Bight (MAB) and Georges Bank during the spring transition from well‐mixed to stratified conditions. Spatial and temporal extents of enhancement vary interannually, and expression is intermittent intraannually. This feature can span the entire MAB and southern flank of Georges Bank (∼1,100 km) and can be expressed for as long as 10 weeks (mid‐April to late June). Pigment concentrations within the feature average more than two times that of adjacent shelf and slope waters. Enhanced shelfbreak chlorophyll consistently coincided with the shelf‐slope front and often extended inshore of the surface outcrop of the front a few to ∼10 km. In all years except 1986, it coincided with seaward entrainment of shelf water by Gulf Stream warm‐core rings (WCRs) or meanders. Shelfbreak chlorophyll enhancement was most pronounced during 1980. Using satellite and in situ observations, we found that during 1980, it coincided with the shelf‐slope front for 10 weeks, and, unlike the spring bloom, it was dominated by the nanophytoplankton (&lt;20 um) size fraction. During the peak of the 1980 occurrence, four WCRs simultaneously interacted with shelf water, and chlorophyll enhancement inshore of one WCR coincided with a slope‐water intrusion onto the shelf. Empirical orthogonal function (EOF) decomposition of CZCS images for late March–June 1980 showed that shelfbreak enhancement was strongly pronounced in an EOF that accounted for &gt;10% of the variance about the mean. This annual biological feature, brought to light in satellite ocean color imagery, is an important aspect of the shelf‐slope ecology of the MAB and Georges Bank.

The onset of a bloom after deep winter convection in the northwestern Mediterranean sea: mesoscale process study with a primitive equation model

The importance of mesoscale processes for primary production predictions is examined in a process study concerning the onset of the spring bloom after deep winter convection in the northwestern Mediterranean sea. Winter deep convection brings nutrient to the enlightened surface layer, but inhibits photosynthesis; phytoplankton biomasses are very low. As soon as restratification occurs, vertical mixing is blocked and a strong bloom onsets. Coastal Zone Color Scanner images have emphasized a strong mesoscale signal in the sea surface chlorophyll during this period. Mesoscale heterogeneity of the mixed-layer depth, due to the baroclinic instabilities associated with the process of deep water formation, is indeed responsible for the mesoscale variability of primary production. To ascertain interactions between hydrological processes and primary production occurring at mesoscales, a primary production model with a parameterization of production inhibition in situations of deep mixing is embedded in a three-dimensional primitive equation model with explicit mixed-layer physics. The model is initialized with a circular chimney of dense water surrounded by a stratified ocean. Two experiments are performed using different treatments of lateral mixing. In the first experiment, the horizontal diffusion is set to a low level so that mesoscale activity can be explicitly resolved. Surface density meanders of 50 km wavelength develop at the periphery of the chimney. These meanders, and the associated vertical motions, induce the sinking and spreading of the chimney, and subsequent surface restratification. Upward motions are responsible for mesoscale mixed layer shallowing, leading to an enhancement of primary production. Maxima of productivity are obtained at the edge of the chimney, where mesoscale activity is the most intense, in agreement with in situ data. In the second experiment, the horizontal diffusion is set to a high level so that lateral mixing occurs primarly through those terms: explicit mesoscale activity is completely damped. The initial structure of the chimney progressively disappears due to the horizontal diffusion of density across the isopycnals instead of three-dimensional redistribution. Mixed-layer depth and productivity are homogeneous. It is shown that instantaneous primary production can be underestimated by a factor of 4 when mesoscale eddies are not explicitly solved. This finding questions the evolution of large-scale coarse resolution climatic models of the oceanic carbon cycle.

Modeling nitrate fluxes in an open coastal environment (Gulf of Lions): Transport versus biogeochemical processes

A 13‐compartment model of primary production and degradation of dissolved organic matter has been coupled with a general circulation model in an open coastal environment (Gulf of Lions, Mediterranean) so as to quantify exchanges with the open sea. The biogeochemical model had been previously calibrated with a one‐dimensional vertical version on a 1‐year data set, and this simulation provides boundary conditions for the three‐dimensional model. After a 1‐year spin‐up simulation, quasi‐equilibrium is obtained for all the compartments, and the results of an additional annual simulation are compared with coastal zone color scanner images and other data collected in the area: the simulation of chlorophyll a, nutrient concentrations, and primary production is quite satisfactory, although the spring bloom starts slightly earlier than usual. Total gross community production is estimated at 76 g C m−2 y−1, and New Gross Community Production at 45 g Carbon m−2 y−1 (f = 0.37). The model was used to understand the nitrate annual cycle in the coastal zone of the Gulf of Lions. Nitrate input from the Rhône river, from the sediment, and from marine advection are compared in terms of potential fertilization. Model results indicate that, regarding the open sea, the margin acts most of the time as a sink for nitrate. However, during winter, when phytoplankton growth is reduced and cascading of dense waters is active along the shelf, the margin is shown to export nitrate toward the open sea. These results may qualitatively apply to other open Mediterranean margins, although the intensity of the fluxes would be modulated by local features.

Chlorophyll estimation in the Southeastern Mediterranean using CZCS images: adaptation of an algorithm and its validation

Abstract Region-specific algorithm to derive chlorophyll concentration from Coastal Zone Color Scanner (CZCS) images was adopted following field measurements in the Southeastern Mediterranean Sea, in June 1992. Reflectance in the range 400 to 700 nm, chlorophyll concentration and seston dry weight were measured simultaneously at 21 stations. The optical model used based on algorithm for Case I waters, using shipboard simulation of CZCS channels for adaptation of parameters. The resulting model was validated with an independent data sets from series of cruises in the Southeastern Mediterranean (1981–1984). Concurrent CZCS images were used for retrieval of the radiance ratio Channel 1/Channel 3, for each station. The resulting ratios were then used to calculate chlorophyll concentrations. A comparison between those calculated concentrations and the actual chlorophyll measurements showed that an estimation error of chlorophyll as low as 0.013 mg m −3 was achieved.

Numerical simulations of the Arabian Sea using tracers as proxies for phytoplankton biomass

A set of numerical simulations were conducted to determine the potential interannual variation of phytoplankton biomass in the Arabian Sea. A four‐layer, reduced gravity model was forced with interannually varying, monthly mean winds for the years 1981 to 1990 and a seasonally varying, monthly mean climatological heat flux. Passive tracers, with fixed source and decay rates, were used as proxies for phytoplankton. The tracers were released in the second layer, and upwelling combined with interlayer mixing transported the tracers into the first (the top) layer of the model. In the top layer advection, diffusion and specified decay rate produced distributions of tracer concentration that we compare with coastal zone color scanner (CZCS) images. In addition to maps of tracer distribution, we constructed time series from areal averages of tracer concentration. The time series are particularly useful for analyzing the temporal variability of the tracer concentrations. There was substantial interannual variability in the results and a strong seasonal signal associated with the southwest monsoon. The years of maximal tracer concentration were 1990 and 1985. The first was associated with a decadal maximum in wind speed and the second with a large wind stress curl over the western Arabian Sea. The years with minimal tracer concentration were 1981 through 1983. We compare our results with published analyses of composited CZCS images during the southwest monsoon in 1981 and 1982. We find general agreement between our results and the CZCS images during the peak of the southwest monsoon; however, we find that during the early phase of the monsoon, when the winds are strengthening rapidly with time, the CZCS images are badly aliased by year to year changes in the time periods used to composite available satellite data.

Time scales of pattern evolution from cross‐spectrum analysis of advanced very high resolution radiometer and coastal zone color scanner imagery

We have selected square subareas (110 km on a side) from coastal zone color scanner (CZCS) and advanced very high resolution radiometer (AVHRR) images for 1981 in the California Current region off northern California for which we could identify sequences of cloud‐free data over periods of days to weeks. We applied a two‐dimensional fast Fourier transform to images after median filtering, (x, y) plane removal, and cosine tapering. We formed autospectra and coherence spectra as functions of a scalar wavenumber. Coherence estimates between pairs of images were plotted against time separation between images for several wide wavenumber bands to provide a temporal lagged coherence function. The temporal rate of loss of correlation (decorrelation time scale) in surface patterns provides a measure of the rate of pattern change or evolution as a function of spatial dimension. We found that patterns evolved (or lost correlation) approximately twice as rapidly in upwelling jets as in the “quieter” regions between jets. The rapid evolution of pigment patterns (lifetime of about 1 week or less for scales of 50–100 km) ought to hinder biomass transfer to Zooplankton predators compared with phytoplankton patches that persist for longer times. We found no significant differences between the statistics of CZCS and AVHRR images (spectral shape or rate of decorrelation). In addition, in two of the three areas studied, the peak correlation between AVHRR and CZCS images from the same area occurred at zero lag, indicating that the patterns evolved simultaneously. In the third area, maximum coherence between thermal and pigment patterns occurred when pigment images lagged thermal images by 1–2 days, mirroring the expected lag of high pigment behind low temperatures (and high nutrients) in recently upwelled water. We conclude that in dynamic areas such as coastal upwelling systems, the phytoplankton cells (identified by pigment color patterns) behave largely as passive scalars at the mesoscale and that growth, death, and sinking of phytoplankton collectively play at most a marginal role in determining the spectral statistics of the pigment patterns.

Dynamics of the transition zone in coastal zone color scanner‐sensed ocean color in the North Pacific during oceanographic spring

A transition zone in phytoplankton concentration running across the North Pacific basin at 30° to 40° north latitude corresponds to a basin‐wide front in surface chlorophyll observed in a composite of coastal zone color scanner (CZCS) images for May, June, and July 1979–1986. This transition zone with low chlorophyll to the south and higher chlorophyll to the north can be simulated by a simple model of the concentration of phytoplankton, zooplankton, and dissolved nutrient (nitrate) in the surface mixed layer of the ocean applied to the North Pacific basin for the climatological conditions during oceanographic springtime (May, June, and July). The model is initialized with a 1° × 1° gridded estimate of wintertime (February, March, and April) mixed layer nitrate concentrations calculated from an extensive nutrient database and a similarly gridded mixed layer depth data set. Comparison of model predictions with CZCS data provides a means of evaluating the dynamics of the transition zone. We conclude that in the North Pacific, away from major boundary currents and coastal upwelling zones, wintertime vertical mixing determines the total nutrient available to the plankton ecosystem in the spring. The transition zone seen in basin‐scale CZCS images is a reflection of the geographic variation in the wintertime mixed layer depth and the nitracline, leading to a latitudinal gradient in phytoplankton chlorophyll.

A numerical analysis of shipboard and coastal zone color scanner time series of new production within Gulf Stream cyclonic eddies in the South Atlantic Bight

Eddy‐induced upwelling occurs along the western edge of the Gulf Stream between Cape Canaveral, Florida, and Cape Hatteras, North Carolina, in the South Atlantic Bight (SAB). Coastal zone color scanner images of 1‐km resolution spanning the period April 13–21, 1979, were processed to examine these eddy features in relation to concurrent shipboard and current/temperature measurements at moored arrays. A quasi‐one‐dimensional (z), time‐dependent biological model, using only nitrate as a nutrient source, has been combined with a three‐dimensional physical model in an attempt to replicate the observed phytoplankton field at the northward edge of an eddy. The model is applicable only to the SAB south of the Charleston Bump, at ∼31.5°N, since no feature analogous to the bump exists in the model bathymetry. The modeled chlorophyll, nitrate, and primary production fields of the euphotic zone are very similar to those obtained from the satellite and shipboard data at the leading edges of the observed eddies south of the Charleston Bump. The horizontal and vertical simulated fluxes of nitrate and chlorophyll show that only ∼10% of the upwelled nitrate is utilized by the phytoplankton of the modeled grid box on the northern edge of the cyclone, while ∼75% is lost horizontally, with the remainder still in the euphotic zone after the 10‐day period of the model. Loss of chlorophyll due to sinking is very small in this strong upwelling region of the cyclone. The model is relatively insensitive to variations in the sinking parameterization and the external nitrate and chlorophyll fields but is very sensitive to a reduction of the maximum potential growth rate to half that measured. Given the success of this model in simulating the new production of the selected upwelling region, other upwelling regions for which measurements or successful models of physical and biological quantities and rates exist could be modeled similarly.

Minimizing systematic errors from atmospheric multiple scattering and satellite viewing geometry in coastal zone color scanner level IIA imagery

Water‐leaving radiances and phytoplankton pigment concentrations are calculated from coastal zone color scanner (CZCS) radiance measurements by removing atmospheric Rayleigh and aerosol radiances from the total radiance signal measured at the satellite. The single greatest source of error in CZCS atmospheric correction algorithms is the assumption that these Rayleigh and aerosol radiances are separable. Multiple‐scattering interactions between Rayleigh and aerosol components cause systematic errors in calculated aerosol radiances, and the magnitude of these errors is dependent on aerosol type and optical depth and on satellite viewing geometry. A technique was developed which extends the results of previous radiative transfer modeling by Gordon and Castaño to predict the magnitude of these systematic errors for simulated CZCS orbital passes in which the ocean is viewed through a modeled, physically realistic atmosphere. The simulated image mathematically duplicates the exact satellite, Sun, and pixel locations of an actual CZCS image. Errors in the aerosol radiance at 443 nm are calculated for a range of aerosol optical depths. When pixels in the simulated image exceed an error threshold, the corresponding pixels in the actual CZCS image are flagged and excluded from further analysis or from use in image compositing or compilation of pigment concentration databases. Studies based on time series analyses or compositing of CZCS imagery which do not address Rayleigh‐aerosol multiple scattering should be interpreted cautiously, since the fundamental assumption used in their atmospheric correction algorithm is flawed.

Interpretation of the coastal zone color scanner signature of the Orinoco River plume

The Caribbean Sea is an area that traditionally has been considered oligotrophic, even though the Orinoco River contributes large quantities of fresh water, nutrients, and other dissolved materials to this region during the wet boreal (fall) season. Little is known about the impact of this seasonal river plume, which extends from Venezuela to Puerto Rico shortly after maximum discharge. Here we present results from a study of the bio‐optical characteristics of the Orinoco River plume during the rainy season. The objective was to determine whether the coastal zone color scanner (CZCS) and the follow‐on sea‐viewing wide‐field‐of‐view sensor (SeaWiFS) satellite instrument can be used to assess the concentrations of substances in large river plumes. Recent in situ shipboard measurements were compared to values from representative historical CZCS images using established bio‐optical models. Our goal was to deconvolve the signatures of colored dissolved organic carbon and phytoplankton pigments within satellite images of the Orinoco River plume. We conclude that the models may be used for case II waters and that as much as 50% of the remotely sensed chlorophyll biomass within the plume is an artifact due to the presence of dissolved organic carbon. Dissolved organic carbon originates from a number of sources, including decay of dead organisms, humic materials from the soil, and gelbstoff.

North Brazil Current retroflection eddies

During 1989–1992, six different anticyclonic eddies were observed to translate up the coast of South America between 7°N and 12°N. These eddies, which are similar to those recently observed with coastal zone color scanner images, current meters, and altimetry, are inferred to have formed from pieces of the North Brazil Current, which retroflects or veers offshore near 7°N to flow eastward. The maximum near‐surface diameter of the eddies when newly formed is estimated to be 400 km. The retroflection eddies were identified in trajectories of four surface drifters and three SOFAR floats at depths from 900–1200 m that looped as they translated northwestward with a mean velocity of 9 cm/s. The two longest trajectories were (1) by a 900‐m float beginning near 7°N in August 1989 that looped 14 times over 151 days with a maximum diameter of 140 km and (2) by a surface drifter beginning near 7°N in October 1990 that looped 9 times over 116 days with a maximum diameter of 250 km and maximum swirl speed of 80 cm/s at that diameter. The mean rotation periods of these two eddies were 11 and 13 days, respectively, with the rotation period of small loops at 900 m around 7 days. Three of the eddies were observed during the period August 1989 to April 1990. Retroflection eddies are considered to provide significant northward volume transport, ∼3 Sv, along the western boundary even during the months when the North Brazil Current has been observed to retroflect into the countercurrent.

Satellite‐derived phytoplankton pigment structures in the Portuguese upwelling area

A series of 25 Nimbus 7 coastal zone color scanner images obtained from July 1981 to September 1983 in Portuguese coastal waters was analyzed in order to study the space‐time distribution of phytoplankton pigment. In the images from winter and spring, the phytoplankton distributions showed no significant spatial patterns. During the upwelling season, from late June to October, several recurrent patterns were observed. Filaments of high pigment concentration extended zonally 200 km off the west coast of Portugal, north of Lisbon. Pigment‐rich plumes were observed south of capes along the west coast, south of Lisbon, and parallel to the south coast of Portugal. The location of the filaments and plumes coincided with topographic features such as submarine ridges. South of Lisbon the location of plumes was also related to coastal morphology. The temporal variability of the pigment patterns was compared with the wind‐induced offshore Ekman transport calculated from measurements made at two meteorological stations, located at Lisbon and Faro, on the south coast. A significant relationship was found: well‐developed phytoplankton structures were generally related to moderate or intense offshore transport, whereas the absence of plumes corresponded to either weak offshore transport or coastal convergence. For each of the three coastal areas in this study, scatter plots showed linear relationships between digital counts of phytoplankton concentration and sea surface temperature. The spatial variation of the temperature range decreased in time after the onset of an upwelling event. One cause of this variation could be the north‐south change in temperature of the upwelled Eastern North Atlantic Central Water along the Portuguese coast.

On the seasonal phytoplankton concentration and sea surface temperature cycles of the Gulf of Mexico as determined by satellites

Monthly climatologies of near‐surface phytoplankton pigment concentration and sea surface temperature (SST) were derived for the Gulf of Mexico from multiyear series of coastal zone color scanner (CZCS) (November 1978 to November 1985) and advanced very high resolution radiometer (AVHRR) (January 1983 to December 1987) images. We complement these series with SST from the comprehensive ocean‐atmosphere data set (1946–1987) and Climate Analysis Center (1982–1990), and hydrographic profile data from the NOAA National Oceanographic Data Center (1914–1985). The CZCS ocean color satellite data provide the first climatological time series of phytoplankton concentration for the region. The CZCS images show that seasonal variation in pigment concentration seaward of the shelf is synchronous throughout the gulf, with highest values (&gt;0.18 mg m−3) in December to February and lowest values (∼0.06 mg m−3) in May to July. Variation in SST is also synchronous throughout the gulf, with maxima in July to September and minima in February to March, The amplitude of the SST variation in the western gulf is about twice that observed in the eastern gulf, and SST maxima and minima persist longer in the west. Larger amplitudes in SST variation are also observed toward the margins. While annual cycles of SST and pigment concentrations are out of phase relative to each other, the phases of mixed layer depth change and pigment concentration change are similar. Model simulations suggest that the single most important factor controlling the seasonal cycle in surface pigment concentration is the depth of the mixed layer. The combined use of ocean color and infrared images permits year‐round observation of spatial structure of the surface circulation in the gulf and the pattern of dispersal of the Mississippi River plume. Infrared images are most useful between November and mid‐May, when strong SST gradients occur. During this time, pigment concentrations are high and can be horizontally homogeneous. In contrast, between late May and October, SST fields are uniform, but the Loop Current and large anticyclonic eddies could be traced with the CZCS. Three anticyclonic eddies were observed in 1979, and at least two were observed in 1980. No eddies were observed during summers of subsequent years in the CZCS time series, but this may be a result of the dramatic decrease in the satellite sampling rate. The series of color images showed that small parcels of Mississippi River water were frequently (2–4 times a year) entrained in the cyclonic edge of the Loop Current, stretched along the Current, and carried to the southeast along the western Florida shelf. However, most of the Mississippi River water flowed to the west, following the Louisiana‐Texas coast as far south as the Mexico‐United States border. Here, a persistent cyclone may reside, exporting shelf constituents to deeper regions of the gulf.

Coupling of coastal zone color scanner data to a physical‐biological model of the southeastern U.S. continental shelf ecosystem: 1. CZCS data description and Lagrangian particle tracing experiments

Nine Coastal Zone Color Scanner (CZCS) images from the southeastern U.S. continental shelf area during April 1980 were processed, and the resulting chlorophyll distributions were analyzed in conjunction with concurrent flow and temperature fields which were obtained from an optimal interpolation of current meter measurements from 17m. The chlorophyll distributions obtained from the CZCS showed high chlorophyll concentrations over the shelf and low chlorophyll concentrations over the offshore shelf region. The variance of the CZCS‐derived chlorophyll distributions was highest at the shelf break region. The chlorophyll patterns at the shelf break varied with time and space and showed shapes characteristic of typical Gulf Stream frontal eddies, high‐chlorophyll bands that were oriented along‐shelf, isolated high‐chlorophyll patches, and simple waveforms. Each event observed in the chlorophyll patterns corresponded with a Gulf Stream frontal eddy event that could be identified in the optimally interpolated flow and temperature fields. Lagrangian particle tracing experiments were also conducted to track the movement of the features observed in the CZCS images. The particles traced in these experiments showed movement that was consistent with the evolution of the patterns observed in the CZCS chlorophyll distributions. Results of the Lagrangian experiments indicate that the optimal interpolation of the current meter measurements reproduced the flow between successive CZCS images. These experiments provide evidence that the high‐chlorophyll bands seen in the CZCS chlorophyll distributions were produced by the passing of Gulf Stream frontal eddy events.

Empirical orthogonal function analysis of cloud‐containing coastal zone color scanner images of northeastern North American coastal waters

Empirical orthogonal function (EOF) analyses were performed on 36 coastal zone color scanner (CZCS) images of the Mid‐Atlantic Bight and Gulf of Maine taken from February 28, 1979, through July 9, 1979. The EOF procedure was modified to allow images with significant portions of data missing because of clouds to be included in the analysis. Analyses performed both with the temporal and with the spatial means removed from the data produced similar results. A strong gradient existed in the temporal mean from high pigment concentrations nearshore to much lower pigment concentrations offshore. The first mode of the temporally averaged analysis explained 21.5% of the variance and showed a south‐to‐north pattern corresponding to an April Mid‐Atlantic Bight bloom and a June bloom over Nantucket Shoals and Platts Bank. The second temporally averaged mode explained 18.3% of the variance and corresponded to a shelf‐wide phytoplankton bloom that resulted in increased pigment concentrations through the spring. The first mode of the spatially averaged analysis explained 59.1% of the variance and had a pattern similar to the temporal mean. The second mode explained an additional 13.2% of the variance and was almost identical to the first mode in the temporally averaged analysis. The inclusion of cloud‐containing images introduced some small‐scale spatial noise into the EOF analyses. However, the inclusion of these images in the EOF analysis allowed postanalysis image reconstruction that provided estimates of pigment concentrations beneath the clouds.