J. Exp. Mar. Biol. Ecoi., 1990, Vol. 143, pp. 139-146 139 Elsevier JEMBE 01498 Observations on double chlorophyll maxima in the vicinity of the Fraser River plume, Strait of Georgia, British Columbia W. P. Cochian ‘, P. J. Harrison ‘v2, P. J. Clifford ’ and K. Yin 1 ‘Department zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA of Oceanography, 2Department of Botany, University of British Columbia, Vancouver. B&s-h Columbia, Canada (Received 15 February 1990; revision received 6 July 1990; accepted 23 July 1990) zyxwvutsrqponmlkjih Abstract: During a cruise to study nutrient and phytoplankton dynamics in the vicinity of the Fraser River plume, vertical profiles of in vivo fluorescence, temperature, salinity, nitrate and silicate were taken. Double chlorophyll maxima at 4-6 and 12-14 m were observed onIy during the late flood tide at two stations on one day and at one station the fallowing day during the 1l-day cruise. Ac~omp~y~g vertical profiles of nitrate and silicate at two stations showed concentration minima at depths corresponding approximately to the two chlorophyll maxima. The upper chlorophyll maximum occurred at the base of the thermochne and halocline and at the beginning of the nitracline, while the lower, chlorophyll maximum was not associated with any features in the temperature or salinity vertical profiles, although minima in nitrate and silicate were associated with it. The most plausible explanation for the observed double chlorophyll maxima is that the upper chlorophyll maximum formed in plume water while the lower chlorophyll maximum originated in waters seaward of the plume and was advected to the plume area during flood tides and possibly entrained into the plume as the freshwater plume flowed seaward. Hence both maxima probably formed due to nitrogen limitation of the surface waters, but at two different depths and flood tides superimposed the two maxima on top of each other due to density differences in the two water masses. These observations suggest that double chlorophyll maxima may only be observed in this riverine plume front during late flood tides and only if a relatively deep (i.e., deeper than the plume maximum) chlorophyll maximum is formed in waters beyond the plume. Other plausible mechanisms of double chlorophyll maxima formation are discussed. Key words: Chlorophyll maximum; Estuarine; Fluorescence; Nutricline; Plume In vivo fluorescence is a standard technique used to estimate ph~opl~kton Chl zyxwvutsrqp a concentration (Lorenzen, 1966) and it is frequently used as an index of ph~op~~kton biomass. The advantage of the in vivo fhrorescence technique in field studies is that it can be used to quickly obtain a vertical profile of Chl a in the euphotic zone by simply lowering a hose through the water column and pumping the profiled water to a shipboard fluorometer. Correspondence address: W. P. Cochlan, Marine Biology Research Division, A-002, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA 92093, USA. 0022-0981~90/$03.50 0 1990 Eisevief Science Publishers B.V. (Biomedical Division) 1411 W.P.COCHLAN ETAL. Subsurface fluorescence maxima are generally Chl u maxima and these subsurface chlorophyll maxima (SCM) are often phytoplankton biomass maxima, but not usually primary productivity maxima (Cullen & Eppley, 1981). A SCM is a common feature of many vertical profiles of Chl a and these SCM have been classified on the basis of their fo~ation (Cullen & Eppley, 198 1; Cullen, 1982). The SCM are observed in most oceans during periods when the water column is stratified, and generally are found in close proximity to the l-5 “/, surface it-radiance level (e.g., Cullen & Eppley, 198 1; Kimor et al.. 1987) and strong vertical gradients including the nitracline, thermocline and pycnocline (Cullen, 1982, and references therein). The often close association between the nitracline and the SCM suggests that the growth of phytoplankton above and below the SCM may be limited by nitrogen and light, respectively. In this paper we report the first observations of two distinct SCM within a common water column with accomp~ying vertical profiles of temperature and salinity, and ambient concentrations of nitrate and silicate. These observations were made during a study of the primary productivity (Harrison et al., submitted) and nitrogenous nutrition (Cochlan et al., in prep.) of phytoplankton in the Strait of Georgia, British Columbia. MATERIALS AND METHODS The cruise was conducted from 20 to 3 1 July 1987 in the Strait of Georgia, British Columbia, in the vicinity of the Fraser River plume during a spring-neap tidal cycle (Fig. 1). The hydrography/circulation (LeBlond, 1983) and biology (Harrison et al., 1983) of the study area have been described previously and a detailed account of our sampling methods and analytical techniques is given in Clifford et al. (1989). Underway measurements of temperature and salinity, determined with an InterOcean 5 14A CSTD equipped with model 5 13D probe, were used to define the riverine plume boundaries and select three main productivity stations: the inner plume (Stn. 3) the mid-plume (Stn. 2) and the outer plume (Stn. 1) (Fig. 1). At these stations continuous vertical profiles (0 to w 20 m) of temperature, salinity, in vivo fluorescence, nitrate plus nitrite and silicate were obtained. A darkened hose (1.5 cm ID) was fixed to the CSTD probe, and as the probe was lowered (z 1 m - min -- ’ ). water was pumped to the ship’s deck with a mRoy FR162-144 diaphragm pump ( a 1 1. min - ‘). Water from the vertical profiles was pumped through a Turner mode1 ill ~uorometer (equipped with a flowthrough cell) to measure in vivo fluorescence, and a Technicon autoanalyser II to measure ambient concentrations of nitrate plus nitrite (Wood et al., 1967) and silicate (Armstrong et al., 1967). These instruments were connected to an analog to digital converter, and the data stored and plotted using an IBM personal computer to give real-time plots of vertical profiles of the parameters (Jones et al., submitted). Pumping time lags and machine analysis delays were measured in order to coordinate sample values from the vertical profnes with the actua1 position in the water column as determined by bottle casts; the depths were generally within + 0.5 m of each other. 141 DOUBLE CHLOROPHYLL MAXIMA. STRAIT OF GEORGIA 5 n I -_ STRAIT \ 1 II -_, I 15 zyxwvutsrqponmlkjihgfedcbaZYXWVU 20 km IO .._ :. : i; OF GEORGIA 48;45 123”40 W t23* 20’ w Fig. 1, Map of study area showing Fraser River plume in southern portion of Strait of Georgia, and three main stations (Stn. 1, outer plume; Stn. 2, mid-plume; Stn. 3, inner plume). Inset shows study area in relation to southern coastline of British Columbia. A sample for phytoplankton species composition was collected at 12 m at Stn. WP6 from the outlet attached to the pumping system. The sample was preserved in acid Lugol’s solution (Parsons et al., 1984) and stored in the dark. A lo-ml subsample was set&d for 24 h and counted using an inverted microscope at 200 and 500 x magnification (UtermUhl, 1958). 142 W.P.COCHLAN E'I‘AL Subsurface light measurements were determined with a Lambda Instruments LI-185B light meter equipped with a LI-192s underwater quantum sensor (2 rc). RESULTS AND DISCUSSION Water circulation in the Strait of Georgia is strongly affected by daily tides (LeBlond, 1983). During a flood tide, water from the Pacific Ocean enters the Strait through the southern entrance and moves up the east side due to the Coriolis effect. On an ebb tide, water moves down the west side of the Strait and out to the ocean. Therefore, the Fraser River plume moves out into the Strait and northwards during a flood tide, and out and southwards during an ebb tide. The size and position of the plume is also affected by spring and neap tides on a fortn~ghtly cycle. The plume is largest during spring tides and may easily reach Stn. 1, depending on the volume of the river discharge (maximum in June) and wind velocity. During the neap tide the plume is much smaller (Z 50?, reduction) and frequently reaches only to Stn. 2 during late July. The zone between the spring and neap tide extent of the plume is referred to as the estuarine plume, and this zone has the highest productivity due to nutrient input from the river and nutrient entr~nment (Harrison et al., submitted). The zone from the river to the boundary formed by the plume on a neap tide is lower in primary pr~u~tjvity, since it is always covered by the turbid, riverine plume. During the post-spring portion of a spring-neap tidal cycle (29 July 1987) two distinct SCM were observed at 4-6 and 12-14 m (Fig. 2). These maxima were first observed during a series of five vertical profiles conducted over a 5-h period during a transect from the inner to the outer plume. They were most pronounced in the outer plume (WP6 and Stn. I), moderately evident within the plume (Stn. 3) and not evident at mid-plume (Stn. 2 and WP7). Accompanying vertical profties of nitrate and silicate at two stations (WP6 and Stn. 3) showed minima in their concentrations at depths corresponding to the two SCM (Fig. 2). The upper SCM occurred at the base of the thermocline and halocline (density is primarily salinity-driven and sigma t profiles follow salinity very closely) and at the beginning of the nitracline. The lower SCM was not associated with any features in the temperature and salinity vertical profiles, but coincided with minima in nitrate and silicate concentrations. At Stn. 3 the lower SCM was mainly below the zyxwvutsrqp 1 “/, light depth, while at WP6, both the shallow and deep SCM were largely above the lo:, surface irradiance depth. The fluorescence peaks represent Chl a concentrations of e 1 (Fig. 2A) to 2 mg * m 3 (Fig. 2B). The phytoplankton species composition at WP6 at 12 m was numerically dominated by Mcromonaspusilla (Butcher) Manton et Parke, cryptomonads, Skeletonema costatum (Greville) Cleve, Chrysochromulina spp. and Chaetoceros spp. (in order of decreasing abundance). The total number of cells was 1.7 x lo4 cells -ml- ’ and M. pusilla composed 44% ofthe total number of cells at this depth, while S. Costa&m, Ch~sochromulina spp. and cryptomonads each composed z 13P’,of the cell numbers. However, in terms DOUBLE A CHLOROPHYLL SALINITY 00 5 (%a) IO TEMP 15 20 MAXIMA, STRAIT (“C) 25 NITRATE 30 0 5 143 OF GEORGIA IO (PM 1 15 20 25 5 z k E - IO F kl n 15 20 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA 80 1.0 0 20 40 60 zyxwvutsrqponmlkjihg .2 .4 .6 .8 0 RELATIVE SALINITY 8 oOI 5 20 1 0 FLUORESCENCE (%a) IO ’ .2 RELATIVE TEMP 15 20 l ’ I .4 .6 SILICATE (“C) 25 .8 FLUORESCENCE NITRATE 30 1.0 0. 5 0 IO (PM) (JJM) 15 20 20 SILICATE 25 40 (JIM) Fig. 2. Vertical profiles of temperature (T), salinity (S), in vivo fluorescence (F), nitrate (N) and silicate (Si) at A: Stn. 3 (1900) and B: WP6 (1745) on 29-30 July 1987, respectively, during a late flood tide ofpost-spring tidal cycle. 134 zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA W. P. COCHLAN ET AL. of cell biovolume or biomass, diatoms dominated (e.g., S. costutum, Chaetoceros spp. and Thaiassiosira spp.). This species composition is very similar to the composition reported at the other stations (l-3) over a range of depths (O-10 m; Clifford et al., 1989; Harrison et al., submitted). Because phytoplankton community structure was similar throughout the whole plume area it is likely that the species composition in the upper SCM and lower SCM were similar, although no samples of the upper SCM were taken. The mode(s) of formation of these double SCM are not apparent, however. there are several possibilities based on biological and/or physical mechanisms. A single SCM frequently forms in the plume area, typically at a depth of 4-6 m a few days after nitrogen is exhausted from the surface water (Clifford et al., 1989; Harrison et al., submitted). Therefore it is the lower SCM at 12-14 m that requires explanation. The most plausible explanation is that the deeper SCM represents the SCM from an area beyond the plume which has been advected into the vicinity of the plume. Waters beyond the plume to the south, with its deeper SCM, may be transported into the plume area during a flood tide by advection and perhaps into the surface plume by entrainment as the freshwater plume flows seaward. We observed that during the post-spring portion of the spring-neap tidal cycle, two double SCM occurred at the inner and outer part of the plume but not at the mid-plume stations (WP7 and Stn. 2) near the end of the flood tide (an increase in tidal height of 3.2 m) on the evening of 29 July. There was no apparent reason for the absence of the SCM at these two latter stations. Vertical mixing could not be inferred from the salinity profiles. However, the double SCM were not observed at Stn. 1 during the ebb tide (mornings of 29-30 July), but they were observed again in the evening on 30 July during the latter portion of the flood tide. Further evidence for the interweaving of water masses hypothesis comes from the double minima in nitrate and silicate concentrations, corresponding to the two SCM (Fig. 2). Therefore the two SCM likely represent interweaving water masses with the upper layer indicative of plume water (increased temperature and decreased salinity in the upper 3-4 m) and the lower SCM representative of water beyond the plume. This interleaving of water masses has also been observed in Saanich Inlet which is near our study site (Parslow, unpubl. data). Both maxima probably form at different depths due to nitrogen limitation in the surface waters (Harrison et al., submitted; Cochlan et al.. in prep.), and the flood tides superimpose the SCM on top of each other due to density differences in the two water masses. Previous studies in other parts of the Strait and in nearby Saanich Inlet indicate that SCM usually form at 5-12 m (Harrison et al., 1983; Parsons et al., 1983; Haigh & Taylor, in press). A second possibility is that the double SCM are the result of successive phytoplankton blooms in the same water mass; the deepest SCM represents an older bloom and the shallower SCM represents a more recent bloom. However, this is an unlikely mode of formation in an area such as the plume where vertical mixing and advection are dynamic processes. A third possibility is that the double SCM were originally result of selective zooplankton grazing; single maxima, and are the a layer of zooplankton at 6-10 m may have DOUBLE CHLOROPHYLL MAXIMA, STRAIT OF GEORGIA 145 consumed a portion of the chlorophyll rn~~~ thereby producing a ph~opl~kton minima in this zone. Zoopl~kton have been shown to aggregate ho~zont~y, with maximum abundance at the boundary of the plume (Parsons et al., 1969; Mackas & Louttit, 1988; Harrison et al., submitted), but it is not known whether they aggregate vertically in such discrete layers. A fourth possibility is that the observed double SCM are a sampling artifact, caused by the movement of an internal wave through the water column while we were lowering our sampling pump. If an internal wave had vertically displaced the water column, it is conceivable that we sampled the same water mass twice at different depths (Holligan, 1978), resulting in apparent double fluorescence maxima (Mackas & Owen, 1982). However, in addition to displacing the fluorescence signal, we would also expect the salinity and temperature signals to be similarly displaced. Plots of sigma t vs. depth (not shown) do not display any vertical displacement of the density structure, and we conclude that the double SCM observed are not due to sampling artifacts. The last plausible mech~ism of fo~ation of double SCM is by verticahy boating dinofla~ellates or flagellates. In this case one might expect to find mainly diatoms (or any non-migrating ph~opl~kter) in one SCM layer and migrating phytoplankton in the other SCM. Our species composition data suggests that the double SCM did not form by such a biological mechanism. We took over 40 vertical profiles during July 1987 in the vicinity of the Fraser River plume and only observed the double SCM on 2 days at three stations (Clifford et al., 1989). However, almost all of these profiles were taken at times other than during the late flood tide. During a cruise in this region during June 1989, we again observed double SCM on three days at two stations; the double SCM at Stn. 3 (inner plume) occurred at the end of the flood tide. These observations suggest that the double SCM may only be found during the late flood portion of post-spring or neap tides, and support our first theory of double SCM formation in which a deeper SCM, from outside of the rive&e plume area, is tidally advected into the plume area and beneath a shallower plume SCM. ACKNOWLEDGEMENTS This research was funded by the Natural Sciences and Engineering Research Council of Canada (NSERC) and a University of British Columbia Graduate Fellowship to W. P. Cochlan. We acknowledge the technical assistance of D. Jones (computer programming and data logging), R. Haigh (phytoplankton counts), and the captain and crew of the C. S. S. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Vector. REFERENCES Armstrong, F. A.J., C. R. Stearns & J. D. H. Strickland, 1967.The measurement of upwelliRgand subsequent biological processes by means of the Technicon Autoanalyzer and associated equipment zyxwvutsrqponmlkjih Deep Sea Res., Vol. 14, pp. 381-389. 136 W. P. COCHLAN ET AL Clifford, P. J., W. P. Cochlan, P.J. Harrison, K. Yin, M.J. Sibbald, L. J. Albright, M. A. St. John & P. zyxwvutsrqp A. Thompson, 1989. Plankton production and dynamics in the Fraser River plume, 1987. Dept. Oceanography, University of British Columbia, Vancouver, MS Rep., No. 51, 1IO pp. Cullen, J. J., 1982.The deep chbrophyll maximum: comparing vertical profiles ofchlorophyii a. Can. J. F&h. zyxwvutsrq Aquat. Sci.. Vol. 39, pp. 791-803. Cullen, J.J. & R.W. Eppley, 1981. Chlorophyll maximum layers of the Southern California Bight and possible mechanisms of their formation and m~nten~ce. Oceand Acta, Vol. 4, pp. 23-32. Haigh, R. & F. J. R. Taylor, in press. Mosaicism of microplankton communities in the northern Strait of Georgia, British Columbia. Mar. Biol. Harrison, P. J., J. D. Fulton, F. J. R. Taylor & T. R. Parsons, 1983. Review of the biological oceanography of the Strait of Georgia: pelagic environment. Can. J. Fish. Aquat. Sci., Vol. 40, pp. 1064-1094. Holligan, P.M. 1978. Patchiness in subsurface phytoplankton populations on the northwest European continental shelf. In, Spatialpattern in plankton communities, edited by J. H. Steele, Plenum Press, New York, pp. 221-238. Kimor. B., T. Berman & A. Schneller, 1987. Phytoplankton assemblages in the deep chlorophyll maximum layers off the Mediterranean coast of Israel. J. Plankton Res., Vol. 9, pp. 433-443. LeBlond, P. H., 1983. The Strait of Georgia: functionai anatomy of a coastal sea. Can. J. Fish. Aquat. Sci., Vol. 40, pp. 1033-1063. Lorenzen, C.J., 1966. A method for the c(~nti~uous me~urement ofin v&ochlorophyll concentration. Deep Sea Res., Vol. 13, pp. 223-227. Mackas, D.L. & G.C. Louttit, 1988. Aggregation of the copepod Ne~ala~~piumch~ at the margin of the Fraser River plume in the Strait of Georgia. Bull. Mar. Sci., Vol. 43, pp. 810-824. Mackas, D. L. & R. W. Owen, 1982. Temporal and spatial resolution of pump sampling systems. Deep Sea Res., Vol. 29. pp. 883-892. Parsons, T. R., R. I. Perry, E. D. Nutbrown, W. Hsieh 6t C. M. Lalli, 1983. Frontal zone analysis at the mouth of Saanich Inlet, British Columbia, Canada. Nar. Biol., Vol. 73, pp. 1-5. Parsons, T. R., Y. Mita & C. M. Lalli, 1984.A manual of chemicaland btilogical methodsfbrseawater analy sis. Pergamon Press, Oxford, 173 pp. Parsons, T.R., K. Stephens & R.J. LeBrasseur, 1969. Production studies in the Strait of Georgia. Part I. Primary production under the Fraser River plume, February to May, 1967.J. Exp. Mar. Biol. Ecof., Vol. 3, pp. 27-38. Uternl~hl. H., 1958. Zur Ve~ollkommung des qu~titativen Phytoplanktoll Methodik. M itt. Int. Ver. Limnol., Vol. 9, 38 pp. Wood, E. D., F. A. J. Armstrong & F. A. Richards, 1967. Determination of nitrate in seawater by cadmiumcopper reduction to nitrite. J. Mar. Biol. Assoc. U.K.. Vol. 47, pp. 23-31.