Species-dependent variations of the absorption coefficient in the ...

Species-dependent variations of the absorption coefficient in the Gerlache Strait

The spectra were partitioned into total (ar ) and detrital (ad) components using the method of Kishino et al. (1984). In the satellite remote sensing pigment retrieval algorithm, the spectral reflectance R(A) can be expressed as: (equation 1)

Eiuc BRODY, B. GREG MITCHELL, OSMUND HOLM-HANSEN, AND MARIA VERNET Marine Research Division Scripps Institution of Oceanography La Jolla, California 92093-0218

Mitchell and Holm-Hansen (1991) found that the bio-optical model used for satellite remote-sensing of chlorophyll + phaeopigment (Gordon et al. 1983, 1988) in Antarctica underestimates concentrations by approximately a factor of 3. It was hypothesized that the variations in pigment-specific absorption coefficients was the source of observed regional differentiation. The causes of these variations are attributed to the "package effect" caused by different pigment composition, cell size, and intercellular chlorophyll concentration (Sathyendranath et al. 1987; Bricaud et al. 1988) and the relative proportion of detritus and phytoplankton absorption (Mitchell and Kiefer 1988). In 1992, during the Research on Antarctic Ecosystem Rates 3 (RACER3) cruise in the Antarctic, over 163 absorption spectra were collected at various stations and depths in Gerlache Strait and near the ice edge. This data set is used to demonstrate the role of species differences as a source of variation in pigment-specific absorption coefficients. Particles were concentrated by filtration and analyzed for chlorophyll and phaeopigments with a filter fluorometer (Holm-Hansen et al. this issue). Absorption samples were collected on GF/F filters, stored in liquid nitrogen for 6 months, measured with a Perkin-Elmer Lamba 6 spectrophotometer, and analyzed using the quantitative filter technique (Mitchell 1990).

E(X) bb() R() = -= C1* Ed(?.) bb(A)+a(A)

where E(A) is the upwelling irradiance, Ed(A) is the downwelling irradiance, bb(?.) is the backscattering coefficient, a(k) is the absorption coefficient and C1 a constant (Gordon et al. 1983,1988). Much of the variation of the reflectance ratio can be attributed to a() (Morel and Prieur 1977), and a(A) can be further decomposed; (equation 2)

a() = a(k) + a (?.) + a()

where a (?.) is absorption of particles, a(X) is absorption of water and a(A5 is absorption of soluble material (Smith and Baker 1981) Since a is constant for each wavelength and a in the Antarctic i minimal (Mitchell this issue), most of the variation in a(A) can b attributed to a(A)1 which can be further decomposed: (equation 3)

a(A.) = aPh( ?4.) + ad()

where aPhQ) is absorption due to phytoplankton and ad(?) i absorption due to detritus (Morrow et al. 1989). Figure 1 shows the correlation between chlorophyll + phaeo. pigment and a(441) for all samples at different stations an4 depths. The slope of the regression is the pigment specific absorp. tion coefficient for nanometers (a*(441)) . The data show a divergence of a*(441) as chlorophyll + phaeopigment increases. Thi4 divergence seems to clearly distinguish two phytoplankton popu lations characterized by a distinctive a*(441) coefficient.

Cyptomonads _

0.5- 0.5-i I

0.4'-4 ' '0.31

Diatoms 'St.FC58 N .1'

11'

•

5

0.2 • —. Wq ,r

5 10 15 20 25 30

Chlorophyll + Phaeopigrnents (mg rn-)

Figure 1. Correlations between a(441)and chi+phaeo for all RACER3 stations. The slope of the regressions represents a* for distinct populations. The estimated a(441) for all stations where the cryptomonads dominated is 0.046 and 0.017 for all stations where diatoms dominated. The values corresponding to the stations FC42 and FC58 are Indicated by the circles.

160



Wavelength (nm) Figure 2. Comparison of two different 5-meter pigment-specifiC absorption spectra selected from figure 1. For 441 nanometers, a Is high for station FC58 and low for station FC42; the inverse is true for each station in the ultraviolet region (330 nanometers). The two spectra located on the bottom of the graph are the corresponding a4, normalized for chlorophyll + phaeoplgment, which is very low, indicating ad is almost insignificant. ANTARCTIC JOURNAL

a(441) [m1] * 100

a(441) [m 2 (mg chI+phaeo1] * 100

-63.8

-64.0

-64.2

-64.4

-64.6

-64.8

-65.0

Figure 3. Grid maps of the RACER3 stations in the Gerlache Strait during the FC leg of the cruise. (a) Contour of a(k); (b) Contour of a*(X). The pattern In figure 3b is similar to the contour of chlorophyll + phaeopigment (Holm-Hansen et al. this issue), but there is an inverse correlation between chlorophyll + phaeopigment and a*. We hypothesize that the observed variations of a* are caused by phytoplankton composition. Based on the floristic analysis (Ferrario, personal communication), samples with high chlorophyll + phaeopigment and low a*p(441) in figure 1 are diatomdominated (diameters greater than 20 micrometers) stations. The stations with high chlorophyll + phaeopigment and high a*(441) are cryptomonad-dominated (diameters less than 5 micrometers) stations. The data support the theory of pigment "package effect" (Morel and Bricaud 1981) as determined by cell size. Large cell phytoplankton such as the diatoms have lower a* (441) (0.017), whereas the small cell phytoplankton such as the cryptomonads have higher a*(441) (0.046). In figure 2, we contrast 5-meter a* spectra from two stations with high chlorophyll + phaeopigment values (14.17 milligrams chlorophyll + phaeopigment per cubic meter for FC42 and 9.60 milligrams per cubic meter for FC58 respectively), and with different a(441) from figure 1. Two very different spectral shapes are evident for stations FC42 and FC58. Station FC58 shows a high a* in the blue (441 nanometer) and a low a* in the ultraviolet (330 nanometers). Station FC42 shows the inverse relationship, low a* in the blue and high a* in the ultraviolet region. Vernet etal. (this issue) found different pigment composition for the stations presented in figure 2, confirming that cryptomonads dominated station FC58 and diatoms dominated station FC42. In addition to having different pigment composition, both samples in figure 2 also reveal a high ratio of accessory pigment to chlorophyll a indicating that the phytoplankton may be lowlight adapted (Jeffrey 1980). Although we do not have detailed

1992 REVIEW

cell size information, we expect high pigment per cell, which could contribute to the pigment "package effect." The variability of a* is important because pigment specific absorption coefficients are used in remote-sensing algorithms (i.e., Frouin et al. this issue). Figure 3a shows the contour of a (441) of FC leg of the RACER3 cruise. This figure is significantly different from the contour of a*p(441) (figure 3b). When comparing a*p(441) with the contour of chlorophyll and phaeopigment (Holm-Hansen etal. this issue), we note that, although the contour pattern is similar, a*p(441) is inversely correlated with chlorophyll + phaeopigment concentration which is consistent with the model of Morel (1988). However, as shown in figure 1, the variation in a* with pigment concentration is not as simple as suggested by Morel (1988). The cell size of phytoplankton species in blooms must also be considered. Variability in a* will affect the estimation of pigment biomass with ocean color sensors such as SeaWiFS and will have to be taken into consideration if we are to have accurate chlorophyll + phaeopigment estimates (Mitchell and Holm-Hansen 1991). We were able to clearly identify two trends in a* for several reasons. One is the clear presence of either very large diatom cells or small cryptomonad cells at bloom stations in Antarctica (HolmHansen and Vernet this issue). Another is the massive bloom in the antarctic coastal waters. In typical oceanographic environments, the small range of chlorophyll + phaeopigment (0 to 5 milligrams per cubic meter) makes it difficult to separate a* for the different phytoplankton. However the large blooms reported in the Antarctic during the spring often show a wide range of chlorophyll +

161

phaeopigment values (0 to 20 milligrams per cubic meter), which makes it easier to distinguish separate a* relationships. The minimal presence of ad is also another important factor in enabling us to find the variation in a* in Antarctica. Figure 2 shows that ad normalized for chlorophyll + phaeopigment was typically negligible. Sosik et al. (this issue) found the ad from RACER3 to be very low compared to the ad from the temperate water off the California coast. Assuming a from equation 2 to be minimal, it will contribute little to variations in a*. Therefore most of the variation is due to a*Ph. Work in progress includes analyzing pigment composition and determining the cell size and phytoplankton composition for the RACER3 cruise. These data should provide us with more information such as pigment per cell and the ratio of accessory pigment to chlorophyll a for all the samples and improve our understanding of particle optics in the Antarctic. We thank the captain and crew of the R/V Polar Duke, Martha Ferrario for the floristic analysis, Chris Fair for technical assistance, and Vincent Spode for graphics. This project was supported by National Science Foundation grant DPP 88-17635 to 0. Holm-Hansen and Maria Vernet. References

Bricaud, A., A. L. Bedhomme, and A. Morel. 1988. Optical properties of diverse phytoplanktonic species: Experimental results and theoretical interpretation. Journal of Plankton Research, 10:851-873. Frouin, R. 1992. Near-surface phytoplankton pigment concentration in the Gerlache Strait derived from aircraft POLDER data. Antarctic Journal of the U.S., this issue. Gordon, H. R. and A. Morel. 1983. Remote Assessment of Ocean Color for Interpretation of Satellite Visiblelmagery: A Review. New York: Springer-

Verlag. Gordon, H. R., 0. B. Brown, R. H. Evans, J . W. Brown, R.C. Smith, K. S. Baker, and D. K. Clark. 1988. A semianalytical radiance model of ocean color. Journal of Geophysical Research, 93:10,909-10,924.

Holm-Hansen, 0. and M. Vernet. 1992. RACER: Distribution, abundance,

A comparison of particulate absorption properties between highand mid-latitude surface waters HEIDI M. S0SIK, MARIA VERNET AND B. GREG MITCHELL

Marine Research Division Scripps Institution of Oceanography University of California, San Diego La Jolla, California 92093-0218

The optical properties of particulate material in the surface waters of the ocean play a critical role in regulating the underwater light field and in determining water-leaving radiance. The

162

and productivity of phytoplankton in Gerlache Strait during austral summer. Antarctic Journal of the U.S., this issue. Jeffrey, S. W. 1980. Algal pigment systems. In P. G. Falkowski (Ed.), Primary Productivity in the Sea. New York: Plenum, 33-58. Kishino, M., C. R. Booth, and N. Okami. 1984. Underwater radiant energy absorbed by phytoplankton, detritus, dissolved organic matter, and pure water. Limnology and Oceanography, 29:340-349. Mitchell, B. G., and D. A. Kiefer. 1988. Variability in pigment specific particulate fluorescence and absorption spectra in the northeastern Pacific Ocean. Deep-Sea Research, 35:665-689. Mitchell, B. G. 1990. Algorithms for determining the absorption coefficient of aquatic particulates using the quantitative filter technique (QFT). Ocean Optics X, SPIE, 1302:137-148. Mitchell, B. G. and 0. Holm-Hansen. 1991. Bio-optical properties of antarctic waters: Differentiation from temperate ocean models. DeepSea Research, 38:1,009-1,028.

Mitchell, B. G. 1992. Predictive bio-optical relationships for polar oceans and marginal ice zones. Journal of Marine Systems, 3:91-105. Morel, A. and A. Bricaud. 1981. Theoretical results concerning light absorption in a discrete medium, and applications to specific absorp tion of phytoplankton. Deep-Sea Research, 28:1,375-1,393. Morel, A. and L. Prieur. 1977. Analysis of variations in ocean color Limnology and Oceanography, 22:709-722.

Morel, A. 1988. Optical modeling of the upper ocean in relation to i4 biogenous matter content (case I waters). Journal of Geophysical Ret search, 93:10,749-10,768.

Morrow, J . H., D. A. Kiefer, and W. S. Chamberlin. 1989. A two-corn ponent description of spectral absorption by marine particles. Limnol ogy and Oceanography, 34:1,500-1,509.

Sathyendranath, S., L. Lazzara, and L. Prieur. 1987. Variations in the spectral values of specific absorption of phytoplankton. Limnology and Oceanography, 32:403-415.

Smith, R. C. and K. S. Baker. 1981. Optical properties of the clearest natural waters (200-800 run). Applied Optics, 20:177-184. Sosik, H. M., M. Vernet, and B. G. Mitchell. 1992. A comparison of particle absorption properties between high- and mid-latitude surface waters. Antarctic Journal of the U.S., this issue. Vernet,M. 1992. RACER: Predominance of cryptomonads and diatoms in the Gerlache Strait. Antarctic Journal of the U.S., this issue.

absorption capability of the phytoplankton fraction of the particulate pool also sets an important limit on primary production. Variability in particulate optical properties thus should be reflected in light propagation models, in algorithms for pigment retrieval from remotely sensed data, and in bio-optical models for primary production. A recent study by Mitchell and HolmHansen (1991) has shown that there are significant differences between high-latitude and temperate ocean waters in pigmentspecific diffuse attenuation coefficients and in the relationship between pigment concentrations and spectral ratios of upwelling radiance. The observed differences have been attributed to variability in particulate optical properties. The goal of this study is to examine differences in measured pigment-specific particulate absorption coefficients between surface waters from the Gerlache Strait and from the California current. This study was conducted on three cruises. Samples were collected from stations in the Gerlache Strait on the Research on Antarctic Coastal Ecosystem Rates 3 (RACER3) cruise in December 1991 to January 1992 and from two California Cooperative Oceanic Fisheries Investigations (Ca1COFI) cruises (9110 and

ANTARCTIC JOURNAL