Journal of Plankton Research Vol.23 no.9 pp.923-938, 2001
© Oxford University Press 2001
Relationship between planktonic dinoflagellate abundance, cysts recovered in sediment traps and environmental factors in the Gullmar Fjord, Sweden
Department Of Marine Botany, Botanical Institute, Göteborg University, Box 461, Se 405 30 Göteborg And 1 Department Of Nuclear Chemistry, Chalmers University Of Technology, Se 412 96 Göteborg, Sweden
 |
Abstract |
|---|
 TOP Abstract INTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
In order to study the relationship between planktonic dinoflagellates, cyst production and environmental factors, a sediment trap study was conducted in the Gullmar Fjord, Swedish west coast, during 21 days in May–June 1998. Five locations for sediment traps were randomly selected every third day. The traps were moored at the five locations and moved to new locations after 3 days. At every location, a CTD depth profile was obtained and water samples were collected for plankton, chlorophyll a and nutrient analysis. Meteorological and hydrographic data for the period were obtained from continuous monitoring. Three dinoflagellate species, which have not previously been recorded from the Kattegat or the Skagerrak (Scrippsiella crystallina, Scrippsiella lachrymosa and Scrippsiella trifida), were encountered during the analysis of cysts from the sediment traps. The abundance of the different species in the motile form encountered in the water column and cyst form encountered in the sediment traps varied greatly. The discrepancy between the number and species encountered in traps and water samples is discussed. No density-dependent relationship between the abundance of planktonic cyst-forming dinoflagellates and the number of cysts recovered could be observed. A multiple regression showed that the variation in cyst yield from the traps for the most abundant species was correlated with water surface temperature, ambient light radiation and the depth of the halocline. The nutrient concentrations (NH4+, NO2–, NO3– and PO43–), which are known to play a crucial role in induction of sexuality and cyst formation under laboratory conditions, correlated poorly with the number of dinoflagellate cysts encountered in the traps.
|
INTRODUCTION |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Ten per cent of the 2000 known dinoflagellate species produce resting cysts (Dale, 1983
). Resting cysts are dormant stages resulting from sexual fusion of gametes, produced in response to certain physiochemical conditions, such as temperature and/or nutrient depletion (von Stosch, 1973). The cysts fall to the bottom and have mandatory dormancy periods of species-specific length, when germination of the cysts is impossible (Pfiester and Anderson, 1987). The cysts constitute a ‘seed bank’ for the region where they are present and they can remain alive in the sediment for several years (Lewis et al., 1999). Many of the harmful species of dinoflagellates, i.e. toxin producers and species causing red tides, form cysts as a part of their life cycle (Matsuoka and Fukuyo, 1995). In order to understand bloom formation and the dinoflagellate life cycle in general, it is important to investigate the transitions between the planktonic stages and the resting stages of the organisms, and the factors influencing these transitions.
Several studies have been performed on cultured algae under laboratory conditions in order to reveal which factors influence encystment of various species of phytoplankton [e.g. (Anderson et al., 1985a; Imai, 1989; Ellegaard et al., 1998)]. The results from these studies show that culture media depleted of nutrients induce sexuality and cyst formation. To some extent, the temperature and light intensity also influence the rate of encystment.
There are few in situ measurements of dinoflagellate cyst production and deposition from the coastal marine environment. Previous studies are from the Baltic Sea (Heiskanen, 1993; Kremp and Heiskanen, 1999), Japanese coastal water (Ishikawa and Taniguchi, 1996) and the Mediterranean Sea (Montresor et al., 1998). These are typically long-term studies where one sediment trap location has been sampled for a long period. The number of cysts deposited in the sediment trap has been compared to the number of planktonic dinoflagellates found in the water over the year and has given valuable information on the seasonal pattern of cyst formation of several species of dinoflagellates.
The problems of quantitative measurements from sediment traps, such as resuspension of bottom material and different efficiency depending on the trap configuration, are well known (Larsson et al., 1986). However, sediment traps are useful for qualitative analysis and relative measurements of the magnitude of cyst formation and deposition in relation to the planktonic community of phytoplankton and environmental factors (Matsuoka et al., 1989).
In this paper, we present a short-term study of dinoflagellate cyst deposition in a fjord located on the Swedish west coast. In addition to planktonic dinoflagellate and cyst enumeration, a wide range of physical, chemical and meteorological parameters were included. The purpose of this study has been to investigate whether there is immediate density dependence between planktonic dinoflagellates and their respective cyst form, and which environmental factors may influence the rate of cyst formation. In order to include the spatial heterogeneity of the parameters analysed, randomly selected sediment trap sites have been used.
|
METHOD |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Research area
The sampling sites were located in the Gullmar Fjord on the Swedish west coast (Figure 1). The fjord has a sill at 45 m depth and a maximum depth of 120 m. Two major current systems affect the coastal area: the low-saline surface Baltic current running parallel to the coast, and the central Skagerrak water circulation pattern resulting in an inflow of more saline North Atlantic water. Hence, the water is stratified in terms of salinity, and a pronounced halocline is present (Rodhe, 1987; Lindahl, 1995). Field studies were conducted during the period 26 May–16 June 1998. Seventy potential sites for posting sediment traps within the research area (58°15'30''–58°18'25''N, 11°26'70''–11°31'85''E) were selected. The selection of these 70 sites was dependent on the depth and existing yacht and ferry routes. Desired depths ranged from 35 to 55 m. Deeper and shallower sites were excluded because the depths of the anchoring sites had to be similar so sediment traps could be moved easily after 3 days. Five positions to post the traps were randomly selected among the 70 potential sites every third day, and every third day the five traps were moved to five new randomly selected positions.
|
Water column measurements
Water samples were collected every third day at the five new randomly selected sites where the sediment traps were to be positioned. Hence, from each sampling site, the data originating from water measurements were collected 3 days prior to the data originating from the material from the sediment traps.
Immediately after reaching the new positions, CTD (conductivity, temperature, depth) profiles were acquired using a CTD (ADM SP C16) with an attached in situ fluorometer (Dr Haardt mini BackScat) and the depths of fluorescence maximum were recorded. Water samples for phytoplankton identification and enumeration, and chlorophyll (Chl) a measurement, were taken with a flexible segmented hose (Lindahl, 1986). The contents of the hose were emptied into a bucket and the water from 0–20 m depth was mixed. From the bucket, 500 ml for phytoplankton enumeration were fixed in 1% acid Lugol's solution (Willén, 1962), 1000 ml were transferred to a polyethylene bottle for studies of live phytoplankton and 500 ml were kept for Chl a analysis. In addition, water samples were collected for nutrient analysis (NO2–, NO3–, NH4+ and PO43–) from the depths of the fluorescence maximum and 5 and 10 m using an all-plastic 1.6 l Ruttner water bottle. A volume of 60 ml from the three depths was collected with syringes and filtered through disposable syringe filters (0.45 µm membrane filter; Sartorius MiniSart NML) into polycarbonate test tubes. The test tubes were stored at –80°C within 4 h, until analysed on a TRAACS 2000 autoanalyser (Grasshoff et al., 1983).
The Chl samples (2 x 100 ml) were filtered on Whatman GF/F glass fibre filters, and frozen at –80°C until analysed. Chlorophyll a was extracted in 7 ml of 90% acetone overnight at –10°C (Jeffrey and Humphrey, 1975). The tubes were turned upside down once before the Chl a concentration was measured with a Turner fluorometer (Turner Designs Model 10-AU). The fluorometer was calibrated using natural phytoplankton samples that had been analysed spectrophotometrically (Jeffrey and Humphrey, 1975).
Sediment trap procedures and dinoflagellate cyst identification
Settled material from the five traps was collected every third day. The collector tubes were removed from the traps and the traps with fresh collector tubes were moved to five new randomly selected positions. Collector tubes were pre-filled with 30 ml of NaCl(s) dissolved in deep-sea water (32 PSU) before mooring in order to prevent resuspension of settled material (Ishikawa and Taniguchi, 1996). No preservatives were added. The traps consisted of a cylindrical plexiglas collector (height 55 cm, diameter 8 cm, aspect ratio 6.9), a flexible tube holder, a wing, a subsurface buoy, a surface marker and an anchoring stone (~40 kg). The rope connecting the anchoring stone to the trap and the subsurface buoy (5 m below the water surface) and a surface marker was adjusted so that the traps were always moored at a depth of 20 m below the surface. We chose a trap depth of 20 m because we wanted to be as low as possible in the photic zone and hence maximize the number of cysts recovered in the sediment traps. Furthermore, we did not want to position the traps too close to the bottom in order to avoid resuspended particles and dinoflagellate cysts from the bottom sediment.
The contents of the trap were emptied onto a 100 µm stainless sieve supported by a 25 µm sieve (Retsch). The tubes were washed out twice with filtered (0.3 µm pore size) deep-sea water (32 PSU). The material on the 100 µm sieve was thoroughly washed with filtered sea water and the remains on the 25 µm sieve were collected into a plastic jar and stored in darkness at a maximum of 8°C until analysed within 2 months. Before microscopic analyses of the sediment trap material, the contents of the plastic jar were emptied into a glass container and sonicated for 5 min. The sonicated material was thereafter transferred to the 25 µm sieve and washed with filtered sea water. The collected material was transferred to a test tube and filled to a total volume of 10 ml with filtered sea water. The slurry was homogenized by shaking the tube, and from this an aliquot of 200 µl was transferred to a new test tube. The new test tube was filled up with filtered sea water to a total volume of 10 ml and poured into an Utermöhl-type sedimentation chamber (Utermöhl, 1958). After a minimum of 8 h, often overnight, the cysts were counted and identified at a magnification of x200, x400 and x1000 using a Zeiss Axiovert 135 inverted microscope. The cysts were identified by means of the literature listed in Persson et al. (Persson et al., 2000; Table II). Very few empty cysts settled in the sediment traps. The dead cysts that occurred were presumed to be resuspended and were not included in the flux estimates, the tables or the graphs. Photographs were taken with an Olympus OM4 camera and video recorded with a Panasonic F15 S-VHS video camera. For scanning electron microscopy (SEM), material from the sediment traps was filtered onto a 1µm Nuclepore polycarbonate filter and quickly rinsed with distilled water. The filters were glued to aluminium stubs and, after drying, they were sputter coated with 20 nm gold/palladium and examined in a JEOL JSM-T 220A scanning microscope.
|
Phytoplankton
Qualitative observation of live phytoplankton and dinoflagellate cysts found in the water column was conducted in order to support the quantitative enumeration of dinoflagellates performed on fixed water samples. For the study of live material, 1 l of sea water collected with the segmented hose was filtered through a 10 µm Nuclepore polycarbonate filter until nearly dry. The plankton suspension was transferred to an Utermöhl-type sedimentation chamber and analysed within 8 h using a Leitz DMIRB.
Water samples fixed in 1% Lugol's iodine settled in 100 ml sedimentation chambers for 48 h and dinoflagellates were identified and enumerated using a Leitz DMIRB.
Meteorological and hydrographic data
The data on atmospheric pressure, precipitation, photosynthetically active radiation (PAR), UV radiation (UVR), air temperature, water surface level, direction and velocity of the wind (Table I) are continuously monitored on a regular basis at Kristineberg Marine Research Station (KMRS) and have been acquired from this monitoring programme. Water salinity and temperature are measured continuously at surface level (by a surface probe at the quay, and at 1 m depth at KMRS) and at 32 m depth in the fjord (58°15'90''N, 11°28'35''E). These data were acquired from the same monitoring programme.
|
Procedures for comparing sediment trap data and plankton records
Plankton samples for quantitative enumeration of motile dinoflagellates were sampled at each of the five randomly chosen stations before the sediment traps were moored. The study covers seven 3-day periods, each from five randomly selected stations. Hence, on 35 occasions, the abundance of motile dinoflagellates (cells l–1) was compared to the number of cysts (cysts m–2 day–1) received in the sediment trap at the same site 3 days after the plankton sample was collected.
Statistical analysis
One of the statistical difficulties that arises when dealing with the kind of problem described in this paper is what function to choose for the evaluation of the results together with the natural fluctuation of the data. According to Occam's razor theorem (Moody, 1935), one should, when in doubt about which model to use, start with the simplest function possible. In our case, this would be the linear function, equation (1):
 | (1) |
) and biological factors. The biological factors included are the number of Scrippsiella spp./P. dalei motile cells recorded from plankton samples, the total number of planktonic dinoflagellates (all planktonic dinoflagellate species; Table II), and the number of Scrippsiella spp./P. dalei cysts found in the plankton samples. In order to limit the number of parameters entered into the regression (36 parameters and 35 observations), a linear correlation analysis was carried out, correlating the different input parameters to the number of cysts in the traps, and to each other, using the CORREL function in Excel 5.0 (Microsoft). Parameters with an absolute value for the correlation coefficient of <0.2 were considered not to affect cyst formation and were thus omitted from further analysis. Furthermore, it is clear that many of the input variables are strongly correlated, e.g. PAR and UVR. Every such group with high internal correlation was described by only one of the variables in the regression analysis. In order to avoid effects arising from the fact that there are different magnitudes of the different input variables, all variables were standardized (N[0,1] distribution). Second-order interactions were not taken into account why the method adopted may eliminate variables needed for a more complex analysis. However, this is not within the scope of this paper.
For ranking the importance between parameters, a stepwise multiple linear regression was carried out for equation (1) using all the input parameters remaining after the initial screening. The analysis was carried out using the STEPR program (Ekberg, 1999) and references therein based on an algorithm found in Bennet and Franklin (Bennet and Franklin, 1954). As a control, to see that the results were not dependent on the software used, the standardized data were also used in the function REGRESSION, in EXCEL, with the same result.
|
RESULTS |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Dinoflagellate species composition
The most common species recorded from the sediment traps was Pentapharsodinium dalei Indelicato et Loeblich III, followed by Scrippsiella trochoidea (Stein) Loeblich III, Protoceratium reticulatum (Claparède et Lachmann) Bütschli, Scrippsiella lachrymosa Lewis and Scrippsiella crystallina Lewis (Table II). The dominating taxa in the plankton were the cyst-forming species complex Scrippsiella spp./P. dalei, and cells from the genera Heterocapsa, Ceratium and Dinophysis. The planktonic motile cells belonging to the genera Scrippsiella and Pentapharsodinium are very difficult to identify to species level by ordinary light microscopy, and therefore the vegetative cells have been grouped into a species complex referred to as Scrippsiella spp./P. dalei (Table II). Correct identification of the motile cells of these species would require staining and/or SEM (Dale, 1977) of each individual cell, which is beyond the scope of this study. On the contrary, the cysts of the same species are easily identified to species level. Owing to the difficulties in identifying the motile cells of these species and for further investigation concerning the density dependence relationship, the motile cells and the cysts were grouped into the species group Scrippsiella spp./P. dalei.
The species S. crystallina, S. lachrymosa and S. trifida Lewis have not previously been recorded from the Skagerrak or the Kattegat (Figure 2A–C).
|
A moderate number of cysts of the paralytic shellfish toxin (PST)-producing species Alexandrium minutum Halim and Alexandrium tamarense (Lebour) Balech were recorded from the traps, but no cells were recorded from the plankton. The species complex Gonyaulax spinifera Diesing is known to produce several different cyst morphotypes, whereas the vegetative planktonic cells have the same plate formula and plate arrangement (Wall, 1973
). No planktonic cells of the G. spinifera species complex were found in the water column, but several different cyst morphotypes (Bitectatodinium tepikiense, Spiniferites elongatus and Spiniferites sp.) were encountered in the traps. The cyst-forming species Lingulodinium polyedrum (Stein) Dodge was rarely found in plankton and not at all in the sediment traps. The cysts denoted as cf. Protoperidinium in Table II are cyst types that are often referred to as spherical brown protoperidinoid cysts in the literature. The motile stages are heterotrophic and they belong to the order Peridiniales, either to the family Protoperidiniaceae or the Diplopsalis group (Dodge, 1982). The only species recorded from the plankton belonging to this family/group is P. depressum, which is not known to form cysts as part of its life cycle. The group of unidentified dinoflagellates encountered in the plankton community was mainly small (<20 µm) naked dinoflagellates. This group does not preserve well in fixatives and details characteristic of the species are often distorted when cells are fixed (Takayama et al., 1998). The unidentified dinoflagellate cysts recovered from the traps (Table II) were also small (<30 µm) and without any distinct features of the cell wall, like spines and mucus. An accumulation body was present in some specimens, but not in others, which indicates that both autotrophic and heterotrophic species were present within this group. The cysts encountered in the traps were almost exclusively non-excysted, which means that the shape of the archeopyle, a commonly used character for species identification, was not visible. The ratio of dead, empty and intact, seemingly viable, dinoflagellate cysts recovered from the traps was ~1:10 000. Pollen grains were commonly encountered in the traps at the beginning of the study period. At the end of the period, large numbers of Cylindrotheca closterium (Ehrenberg) Lewin et Reimann were encountered in the traps. A small number of living ciliates and empty loricas were encountered while examining the contents of a trap on one occasion. We assume that large predators, such as copepods, were removed when the material from the traps was initially sieved (100 µm). The small proportion of empty dinoflagellate cysts and other dead material encountered in the traps indicated that the amount of resuspended material collected in the traps was limited.
Cysts and phytoplankton
In order to examine whether there was any densitydependent relationship between the planktonic motile cells and the number of cysts encountered in the traps of the same species group or species, graphs are presented for the dominant species group, Scrippsiella spp./P. dalei (Figure 3A), and Protoceratium reticulatum (Figure 3B). The numbers of motile planktonic cells (cells l–1) were plotted against the number of cysts (cysts m–2 day–1) encountered in the traps 3 days later from the same site. Neither for Scrippsiella spp./P. dalei nor for P. reticulatum was there a linear correlation between the abundance of planktonic cells and the number of cysts encountered in the traps. On the contrary, when the number of cysts recorded in the traps was at a peak, the number of planktonic cells was rather low. The average number (n = 5) of planktonic Scrippsiella spp./P. dalei or P. reticulatum motile cells plotted against the average number of cysts encountered in the traps of the same species did not display any apparent relationship either (data not shown).
|
In Figure 4
, the average number of planktonic cells recorded in five water samples and the average number of cysts encountered in five traps over each period are depicted. For all values, there are large variations between the average number of planktonic cells and cysts in the five randomly selected sites at each time. Nevertheless, there are some trends in the relationship of planktonic cells versus cysts formed during the period of this study. The average number of planktonic cells of the species group Scrippsiella spp./P. dalei (Figure 4A) declined from day 0 and reached a minimum on day 12, but rose again to a few thousand cells per litre until the end of the period. On the contrary, the number of cysts reached a maximum, up to 1.1 x 106 cysts formed per day and per square metre, during the period from day 9 to 12, and thereafter the number of cysts declined until the end of the period, i.e. day 21. The numbers of vegetative cells of Protoceratium reticulatum (Figure 4B) increased from day 0 to day 3 and then steadily declined over the period. The average number of cysts was high until day 12, and then declined markedly towards the second half of the study period. Figure 4C depicts the total average (n = 5) abundance of cyst-forming motile dinoflagellate cells found among the plankton and the total number of cysts encountered in the traps. This graph resembles the graph displayed for Scrippsiella spp./P. dalei (Figure 4A) due to the fact that this species group was the dominating cyst former. The number of cyst-forming planktonic dinoflagellates was highest during the first period (days 0–3), declined and reached a minimum at day 12, and then increased, although not to the same extent, towards the end of the period. On the contrary, the cysts reached an absolute maximum during the period between days 9 and 12.
|
Environmental factors
The period just before the peak of cyst abundance (days 6–9), and the following period (days 9–12) when the cyst yield from the traps was greatest, were meteorologically unstable days with oscillations of hydrographic andmeteorological parameters. Easterly winds with velocities between 0 and 6 m s–1, atmospheric pressure at a minimum of 1002 mbar, moderate precipitation and a rise in water level were characteristic of the period between days 6 and 9. Thereafter (days 9–12), there followed a period of westerly winds with a velocity up to 12 m s–1, atmospheric pressure at a maximum of 1020 mbar, a decline in water level and no precipitation at all (data not shown). There was a fall in surface salinity from day 6 and onwards (Figure 5A). Water surface temperature was steadily increasing over the 21 day period, which is a normal trend of the early summer season (Figure 5B). The Chl a concentration of the surface water was low, 1–4 µg l–1, during the period of the study (Figure 5C), which is a normal feature in the Gullmar Fjord during the months of May and June (Water Quality Association of the Bohus Coast, 1990–99). Figure 6 displays temperature versus salinity plots of all CTD data from the surface to the bottom at five stations [except day 6 (n = 3) and day 15 (n = 1)] from day 0 to 21. The dots of each graph are clustered in a narrow span, which indicates that the areas investigated (represented by the five randomly selected positions) on each sampling day were homogeneous with respect to hydrographic conditions.
|
|
The NO3– concentration remained low during the whole period of the study. On day 12, the values reached a maximum of 1.7 µM at 10 m depth, but higher concentrations were not present until the very end of this study (Figure 7
).
|
The importance of the environmental factors influencing the rate of cyst production for the dominant species group, Scrippsiella spp./P. dalei, was ranked after performing a multiple regression. The variation in cyst production was tightly connected with the position of the halocline, water surface temperature and ambient light radiation (Table III
). The water surface temperature and PAR were inversely correlated with the number of cysts encountered in the traps. The depth of the halocline was positively correlated with the number of cysts encountered in the sediment traps. Figure 8 shows the relationship between the observed and predicted (by the regression model) number of cysts encountered in the traps. The correlation between the observed and predicted (by the regression model) number of cysts encountered in the traps was 0.79.
|
|
Since the ambient nutrient concentration, especially nitrogen, is known to play a crucial role in induction of sexuality and cyst formation, it should be mentioned that the NH4+, NO2–, NO3– and PO43– concentrations correlated poorly with the number of dinoflagellate cysts encountered in the traps (Table IV
).
|
|
DISCUSSION |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Several studies of dinoflagellate cysts recovered from sediment samples and sediment traps have revealed the presence of dinoflagellate species that have not been recorded earlier from the same area in their planktonic stage (Dale, 1976
; Dodge and Harland, 1991; Montresor et al., 1998). This is also true in this study. The cyst forms of Scrippsiella lachrymosa, S. crystallina and S. trifida (Figure 2A–C) were encountered for the first time in Swedish water. The specimens recorded from the sediment traps in the Gullmar Fjord coincide in morphology with specimens of the same species found in other locations in north-western Europe, i.e. Loch Creran in Scotland (Lewis, 1991), the German Bight in the North Sea and Kiel Bight in the Baltic Sea (Nehring, 1997).
Table II reveals a discrepancy among cyst-forming species of dinoflagellates. Some species were present in the traps, but not in the plankton samples, and vice versa. A moderate number of Alexandrium minutum, A. tamarense and Alexandrium sp. cysts occurred in the sediment traps. Planktonic cells and cysts of Alexandrium sp. were registered in very low numbers during the first qualitative analyses of the live plankton samples, but were not recorded from quantitative fixed plankton samples (Table II). Owing to the small cell size and the difficulties in identifying the species (Taylor et al., 1995), they could have been grouped with the unidentified dinoflagellates (Table II). Low numbers of the cyst-forming species Lingulodinium polyedrum were encountered in the plankton, but no living cysts was found in the sediment traps. Cysts of this species are often the dominant species in recent sediment from the Swedish west coast, but the motile form is known to be most abundant during autumn (Godhe and Wallentinus, 1999; Persson et al., 2000). It is, therefore, not surprising that this species has not been more numerous in the plankton and the traps during this study conducted in May and June. Cysts of Gonyaulax spinifera species complex, Bitectatodinium tepikiense, Spiniferites elongatus and Spiniferites sp., and spherical brown protoperidinoid cyst types, cf. Protoperidinium, were recorded in relatively small numbers. Motile cells of these species could have been overlooked in the plankton samples and grouped with the unidentified dinoflagellates.
The species composition in the plankton samples (Table II) was similar to other years, as stated in monitoring programmes for the Swedish west coast. The dominating dinoflagellate genera during the months of May and June (1990–1998) were Ceratium spp., Dinophysis spp., Heterocapsa spp., Protoperidinium spp., Gymnodinium spp., Gyrodinium spp., Katodinium sp., Scrippsiella spp. and Protoceratium sp. (Edler, 1995; Hernroth and Kuylenstierna, 1998; Godhe and Wallentinus, 1999).
The cyst yield of the sediment traps in this study is similar to other studies conducted previously in marine coastal areas and freshwater lakes. The mean number of cysts encountered in the traps during this study was 3.2 x 105 cysts m–2 day–1, and the maximum yield, 2.7 x 106 cysts m–2 day–1, occurred in trap E on days 9–12. Montresor et al. (Montresor et al., 1998) reported a maximum cyst yield of 1.7 x 106 cysts m–2 day–1 from the coastal Mediterranean, and the maximum cyst yield from Onagawa Bay, Japan, was 3.16 x 105 cysts m–2 day–1 (Ishikawa and Taniguchi, 1996). Studies from a Japanese freshwater lake also display similar cyst yields: 2.3 x 106 cells m–2 day–1 (Park and Hayashi, 1993).
Figure 6 displays the hydrographic conditions, in terms of salinity and temperature, in the studied area from day 0 to day 21. The appearances of the graphs indicate that the overall hydrographic condition in the area is well documented by the five replicate sites investigated on each day. It is, therefore, reasonable to assume that most of the cyst-forming dinoflagellates found in the water column (0–20 m) that actually encysted and formed resting spores sank to the bottom, at least to 20 m depth where the sediment traps were located, within the studied area. With short sampling intervals, every third day, and five replicate traps on each day of investigation spread over a study area as homogeneous as displayed in Figure 6, we assumed that the cysts received in the sediment traps originated from local plankton populations within the fjord. Hence, the problem of transport of cysts out of the area by currents after encystment is minimized, provided that the time interval of 3 days is sufficient for the cysts to sink 20 m, to the depth of the sediment trap. At all sampling sites, on every sampling day, the position of the halocline was positioned above the depth of the sediment trap (Figure 5A). This could entail newly formed cysts being trapped at the level of this density interval. We found numerous cysts in the integrated plankton sample, but since the plankton samples were not collected in a discrete manner, it is impossible to say to what extent this ‘trapping’ occurred. We chose to position the sediment traps at 20 m depth in order to be as low as possible in the photic zone and hence maximize the cyst yield in the traps. Any depth above the halocline would have resulted in uncertainty of catching all cysts formed in the water column, since cyst-forming dinoflagellates are present all through the photic zone. A comparison between the planktonic dinoflagellate population and the number of cysts encountered in the traps was made in order to establish the relationship. Figure 3 gives no support for density-dependent cyst formation. No linear correlation between the number of vegetative planktonic dinoflagellates and the number of cysts recorded from the traps during 3 day intervals could be detected. Observed sinking rates for the calcareous cyst of Scrippsiella trochoidea and the gonaucoloid cyst of Alexandrium tamarense were 11.2 and 9.5 m day–1, respectively, at 22°C and 32 PSU (Anderson et al., 1985b). The lower water temperature during this study will decrease the sinking rates. However, the period of 3 days between plankton sampling and recovery of cysts from the trap ought to be sufficient in order to make such a comparison for the species group Scrippsiella spp./P. dalei and for Protoceratium reticulatum. Previous studies concerning the relationship between phytoplankton and their respective resting stages do vary with respect to density dependence. In several long-term field studies of phytoplankton capable of forming resting stages, it is observed that the maximum numbers of spores or cysts are formed during or just after a period of maximum vegetative cell division (Heaney et al., 1983; McQuoid and Hobson, 1996; Zohary et al., 1998). Observations from a sediment trap study conducted during 4 years in a Japanese lake displayed a nearly continuous recovery of Peridinium bipes f. occulatum (Lindem.) Lef. cysts all year around with a maximum abundance of cysts throughout the blooms (Park and Hayashi, 1993). In the Mediterranean, peaks of Scrippsiella-like cysts followed those of motile peridinoid cells by some weeks (Montresor et al., 1998). Previous encystment studies carried out under laboratory conditions show no clear correlation between plankton cell abundance and cysts formed. No density-dependent induction of sexuality was registered for A. tamarense (Anderson et al., 1984), Gyrodinium uncatenatum Hulburt (Anderson et al., 1985a) or for Peridinium cinctum Ehrenberg and Peridinium willei Huitfeld-Kaas (Chapman and Phiester, 1995).
The most common and successful method to induce sexuality and the formation of cysts or resting stages from vegetative cells in cultures of phytoplankton is to transfer the cells into a nutrient-depleted medium, especially a nitrogen-depleted one (Pfiester and Anderson, 1987; McQuoid and Hobson, 1996). The NO3– and PO43– concentrations in f/2, a common medium used for culturing autotrophic phytoplankton, are 883 and 36.3 µM, respectively (Guillard, 1975). Encystment of Gyrodinium uncatenatum was induced when the cells were transferred from f/2 to f/30 medium (58 µM NO3–) (Anderson et al., 1985a). Chattonella marina (Subrahmanyan) Hara et Chihara encysted when transferred from a nitrogen concentration of 2 mM to 25 µM (Imai, 1989), and Alexandrium tamarense when transferred from f/2 to a nitrogen concentration of 88 µM (Anderson et al., 1984). High cyst yields in sediment traps moored in Lake Kizaki, Japan, of the freshwater-living species P. bipes f. occulatum coincided with low nitrogen levels, 3.5 µM (Park and Hayashi, 1993). Furthermore, several studies have shown that nitrogen starvation is the crucial factor in order to induce sexuality. Other external factors, such as light and temperature, may affect the cyst yield only when the cells are exposed to nitrogen limitation. A mere change in light or temperature will not induce sexuality of the cells (Anderson et al., 1985a; Sako et al., 1987; Park and Hayashi, 1993). The NO3– concentration in the water (5 and 10 m depth) during this study ranged from 0.1 to 3.7 µM. Only at the very end of the investigation period was the NO3– concentration >2 µM (Figure 7). Collected data from 1990–99 showed that the surface water (0–5 m) concentration of inorganic nitrogen in the Gullmar Fjord during spring and summer ranged between 0 and 5 µM, while during the winter months the concentrations never rose above 20 µM (Water Quality Association of the Bohus Coast, 1990–99). Hence, the immediate nutrient history of the fjord, just prior to the time of investigation, ought to be similar to the nutrient situation documented during this study, and the cyst-forming dinoflagellates adapted to low nutrient concentrations. Neither the linear correlation analysis nor the multiple regression analysis indicated that the nitrogen or the phosphorus concentrations were important environmental factors related to the number of cysts formed (Tables III and IV). The multiple regression showed that the most important environmental factors were water surface temperature, solar radiation and the depth of the halocline. This implies, in accordance with earlier studies of laboratory cultures and natural samples in lakes, that the cyst-forming dinoflagellate population was sexually induced during the whole period. The variation in the number of cysts received in the sediment traps depended on the variation in surface water temperature and the intensity of solar radiation. Hence, there was no correlation with the nitrogen or phosphorus data (Table IV), while there was a strong correlation with the data on surface water temperature and solar radiation (Table III). In Figure 8, the standardized data (N[0,1] distribution) on observed and model-predicted numbers of cysts formed are presented. The interpretation of this type of analysis carried out on data collected in such complex systems is difficult. The result that comes out of any multivariate analyses should be treated with caution. Nevertheless, the great number of laboratory studies documented in the literature tempted us to investigate whether the results generated in laboratory experiments could be validated in field studies like this.
Several laboratory studies on dinoflagellate encystment and environmental factors, such as temperature, have revealed a temperature interval when encystment is possible. The optimal temperature in terms of encystment was 21°C for A. tamarense, whereas no cysts were produced above 25°C or below 10°C (Anderson et al., 1984). Cyst production in Gymnodinium nolleri Ellegaard et Moestrup reached a maximum at 22–28°C; no cysts were produced below 13°C or above 33°C (Ellegaard et al., 1998). The freshwater dinoflagellates Peridinium penardii (Lemm.) Lemm and P. bipes f. occulatum had optimal temperatures for encystment at 15 and 15–25°C, respectively (Sako et al., 1987; Park and Hayashi, 1993). In all species, sexuality was induced in nutrient-depleted media, and raising and lowering the temperature thereafter manipulated cyst yields. The surface water temperature during this study fluctuated from 10 to 14°C (Figure 5 B), which are temperatures within the range where encystment has proven successful in cultures of dinoflagellates. It is, therefore, not surprising that water surface temperature falls out as an important environmental factor, affecting the number of cysts formed.
Results from the multiple regression of the dominant species complex Scrippsiella spp./P. dalei showed that PAR was of importance for cyst formation (Table III). The measured parameter UVR correlated strongly with PAR, which is why it has been impossible to separate the two parameters in the statistical analysis. Decreased PAR generated a higher number of cysts recovered from the traps. As for the environmental parameter water temperature, light intensity has proven to be an important factor regulating cyst formation in laboratory cultures of microalgae. However, different species of microalgae seem to respond differently to the effect of irradiance in terms of encystment efficiency. Maximum cyst yield in nutrient-starved cultures of Chattonella marina was found at low light intensities: 
12 µE m–2 s–1 (Imai, 1989). In the freshwater dinoflagellate Wolozynskia apiculata von Stosch, sexual reproduction and encystment were induced when vegetative cells were inoculated in nutrient reduced medium and transferred from 160 to 25 µE m–2 s–1 (von Stosch, 1973). In natural populations of the red-tide-forming, freshwater species Peridinium bipes f. occulatum, encystment has been observed in continuous darkness. However, the encystment rate increased from 1 to 36% of the vegetative population when the light intensity was increased to 105 µE m–2 s–1 (Park and Hayashi, 1993). In a study of P. penardii, the highest cyst yield was found at 120 µE m–2 s–1 and a decreasing cyst yield was documented with lower irradiance (Sako et al., 1987). Some genera of diatoms have also been shown to induce resting stages in response to the ambient light condition (McQuoid and Hobson, 1996).
Previous papers have described damaging effects and reduced growth of phytoplankton and microphytobenthic communities due to enhanced UVR [e.g. (Quesada et al., 1995; Villafãne et al., 1995)]. This study in the Gullmar Fjord indicates that UVR affects the number of cysts formed in a negative manner (Table III). We are not aware of any study where the rate of dinoflagellate encystment has been studied in relation to UVR. However, UVR has been shown to speed up the formation of resting spores in diatom species of the genus Chaetoceros (Helbling et al., 1992). Villafãne et al. found a shift towards a more diatom-dominated phytoplankton community structure and less flagellates, including dinoflagellates, when exposed to enhanced UVR (Villafãne et al., 1995). Further studies on dinoflagellate encystment connected to UVR may reveal whether the result acquired here is an effect of decreased dinoflagellate abundance, which would at least in a longer perspective reduce the number of cysts formed, or whether the rate of dinoflagellate encystment is truly negatively affected by increased UVR. Obviously, more effort has to be put into the investigation of how UVR affects the formation of cysts, either as well-controlled laboratory studies or a significantly larger study in natural waters.
The depth of the halocline was, according to the multiple regression analysis, an important environmental factor in relation to the number of Scrippsiella spp./P. dalei cysts recovered from the traps. It was positively correlated with the number of cysts formed. If we assume that a larger proportion of the cyst-forming dinoflagellates were present at the depth of the halocline, and if the halocline was situated at greater depth, the water temperature and the irradiance would be lower. This is in accordance with the inversely correlated environmental factors, water surface temperature and PAR, on cyst formation.
This study has shown that there was no densitydependent cyst formation over such a short period as 3 days to be detected from quantitative plankton enumeration and sediment trap analyses. The ranking of the importance of environmental factors influencing cyst production revealed that water temperature, ambient light conditions and the position of the halocline were correlated with the cyst yield of the traps. The nutrient concentration, known to be an important factor for dinoflagellate encystment in laboratory cultures, had no correlation with cyst production in this field study.
|
Acknowledgments |
|---|
The authors wish to thank Tyri Asklöf, Johan Burman, Frank Linares and Anette Norén for help during the field part of this study. We also wish to thank the staff at KMRS and Monica Appelgren at Department of Marine Botany, Jakob Hagberg for comments on the statistical part, Dr Melissa McQuoid and Professor Inger Wallentinus for giving valuable comments on the manuscript, and Ernst-Gunnar Persson and Barbro Godhe-Persson for commenting on the language. Three anonymous reviewers are thanked for constructive criticism of the manuscript. This study was supported by grants from Kapten Carl Stenholms Donationsfond and Stiftelsen Birgit och Birger Wåhlströms Minnesfond.
|
REFERENCES |
|---|
|
TOPAbstractINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Anderson, D. M., Kulis, D. M. and Binder, B. J. (1984) Sexuality and cyst formation in the dinoflagellate Gonyaulax tamarensis: cyst yield in batch culture. J. Phycol., 20, 418–425.[ISI]
Anderson, D. M., Coats, D. W. and Tyler, M. A. (1985a) Encystment of the dinoflagellate Gyrodinium uncatenatum: temperature and nutrient effects. J. Phycol., 21, 200–206.[ISI]
Anderson, D. M., Lively, J. J., Reardon, E. M. and Price, C. A. (1985b) Sinking characteristics of dinoflagellate cysts. Limnol. Oceanogr., 30, 1000–1009.
Bennet, C. and Franklin, L. (1954) Statistical Analysis in Chemistry and the Chemical Industry. John Wiley and Sons, New York.
Chapman, A. D. and Pfiester, L. A. (1995) The effects of temperature, irradiance, and nitrogen on the encystment and growth of the freshwater dinoflagellates Peridinium cinctum and P. willei in culture (Dinophyceae). J. Phycol., 31, 355–359.[ISI]
Dale, B. (1976) Cyst formation, sedimentation and preservation: factors affecting dinoflagellate assemblages in recent sediments from Trondheimsfjord, Norway. Rev. Palaeobot. Palynol., 22, 39–60.
Dale, B. (1977) New observations on Peridinium faeroense Paulsen (1905), and classification of small orthoperidinioid dinoflagellates. Br. Phycol. J., 12, 241–253.
Dale, B. (1983) Dinoflagellate resting cysts: ‘benthic plankton’. In Fryxell, G. A. (ed.), Survival Strategies of the Algae. Cambridge University Press, Cambridge, pp. 69–136.
Dodge, J. D. (1982) Marine Dinoflagellates of the British Isles. Her Majesty's Stationery Office, London.
Dodge, J. and Harland, R. (1991) The distribution of planktonic dinoflagellates and their cysts in the eastern and northeastern Atlantic Ocean. N. Phytol., 118, 593–603.
Edler, L. (1995) Fytoplankton Årsrapport 1995. Report to The Water Quality Association of the Bohus Coast, Udevalla (in Swedish).
Ekberg, C. (1999) Uncertainties in actinide solubility calculations using the Th-OH-PO4 system. PhD Thesis, Chalmers University of Technology, Department of Nuclear Chemistry, Göteborg.
Ellegaard, M., Kulis, D. M. and Anderson, D. M. (1998) Cysts of Danish Gymnodinium nolleri Ellegaard et Moestrup sp. ined. (Dinophyceae): Studies on encystment, excystment and toxicity. J. Plankton Res., 20, 1743–1755.
Godhe, A. and Wallentinus, I. (1999) The ports in the Stenungsund area, west coast of Sweden. In Gollasch, S. and Leppäkoski, E. (eds), Initial Risk Assessment of Alien Species in Nordic Coastal Waters. Nordic Council of Ministers, Copenhagen, pp. 140–184.
Grasshoff, K., Erhardt, M. and Kremling, K. (1983) Methods of Seawater Analyses. Verlag Chemie, Weinheim.
Guillard, R. (1975) Culture of phytoplankton for feeding marine invertebrates. In Smith, W. and Chanley, M. (eds), Culture of Marine Invertebrate Animals. Plenum Press, New York, pp. 29–60.
Heaney, S., Chapman, D. and Morison, H. (1983) The role of the cyst stage of the dinoflagellate Ceratium hirundella within a small productive lake. Br. Phycol. J., 18, 47–49.
Heiskanen, A.-S. (1993) Mass encystment and sinking of dinoflagellates during a spring bloom. Mar. Biol., 116, 161–167.
Helbling, E., Villfãne, V., Ferrairo, M. and Holm-Hansen, O. (1992) Impact of natural ultraviolet radiation on rates of photosyntesis and on specific marine phytoplankton species. Mar. Ecol. Prog. Ser., 80, 89–100.
Hernroth, B. and Kuylenstierna, M. (1998) Fytoplankton Analyser 1998. Report to The Water Quality Association of the Bohus Coast, Udevalla (in Swedish).
Imai, I. (1989) Cyst formation of the noxious red tide flagellate Chattonella marina (Raphidophyceae) in culture. Mar Biol., 103, 235–239.
Ishikawa, A. and Taniguchi, A. (1996) Contribution of benthic cysts to the population dynamics of Scrippsiella spp. (Dinophyceae) in Ongawa Bay, northeast Japan. Mar. Ecol. Prog. Ser., 140, 169–178.
Jeffrey, S. and Humphrey, G. (1975) New spectrophotometric equations for determining chlorophylls a, b, c1 and c2 in higher plants, algae and natural phytoplankton. Biochim. Physiol. Pflanzen, 167, 191–194.
Kremp, A. and Heiskanen, A.-S. (1999) Sexuality and cyst formation of the spring-bloom dinoflagellate Scrippsiella hangoei in the coastal northern Baltic Sea. Mar. Biol., 134, 771–777.
Larsson, U., Blomqvist, S. and Abrahamsson, B. (1986) A new sediment trap system. Mar. Ecol. Prog. Ser., 31, 205–207.
Lewis, J. (1991) Cyst–theca relationships in Scrippsiella (Dinophyceae) and related orthoperidinioid genera. Bot. Mar., 34, 91–106.
Lewis, J., Harris, A., Jones, K. and Edmonds, R. (1999) Long-termsurvival of marine planktonic diatoms and dinoflagellates in stored sediment samples. J. Plankton Res., 21, 343–354.
Lindahl, O. (1986) A dividable hose for phytoplankton sampling. ICES, C. M. 1986/L:26, annex 3 (mimeo.).
Lindahl, O. (1995) Long term studies of primary phytoplankton production in the Gullmar fjord, Sweden. In Skjoldal, H. R., Hopkins, C., Erikstad, K. E. and Leinaas, H. P. (eds), Ecology of Fjords and Coastal Waters. Elsevier Science, Amsterdam, pp. 105–112.
Matsuoka, K. and Fukuyo, Y. (1995) Taxonomy of cysts. In Hallegraeff, G. M., Anderson, D. M. and Cembella, A. D. (eds), Manual on Harmful Marine Microalgae. IOC Manuals and Guides No. 33. UNESCO, Paris, pp. 381–401.
Matsuoka, K., Fukuyo, Y. and Anderson, D. M. (1989) Methods for modern dinoflagellate studies. In Okaichi, T., Anderson, D. M. and Nemoto, T. (eds), Red Tides: Biology, Environmental Science, and Toxicology. Elsevier Science, New York, pp. 461–479.
McQuoid, M. R. and Hobson, L. A. (1996) Diatom resting stages. J. Phycol., 32, 889–902.[ISI]
Montresor, M., Zigone, A. and Sarno, D. (1998) Dinoflagellate cyst production at a coastal Mediterreanean site. J. Plankton Res., 20, 2291–2312.
Moody, E. A. (1935) The Logic of William of Ockham. Sheed & Ward, London.
Nehring, S. (1997) Dinoflagellate resting cysts from recent German coastal sediment. Bot. Mar., 40, 307–324.
Park, H.-D. and Hayashi, H. (1993) Role of encystment and excystment of Peridinium bipes f. occulatum (Dinophyceae) in freshwater red tides in Lake Kizaki, Japan. J. Phycol., 29, 435–441.[ISI]
Persson, A., Godhe, A. and Karlson, B. (2000) Dinoflagellate cysts in recent sediments from the west coast of Sweden. Bot. Mar., 43, 69–79.
Pfiester, L. and Anderson, D. M. (1987) Dinoflagellate reproduction. In Taylor, F. (ed.), The Biology of Dinoflagellates. Blackwell Scientific, Oxford, pp. 611–648.
Quesada, A., Mouget, J.-L. and Vincent, W. (1995) Growth of Antarctic cyanobacteria under ultraviolet radiation: UVA concentrations counteracts UVB inhibition. J. Phycol., 31, 242–248.[ISI]
Rodhe, J. (1987) The large scale circulation in the Skagerrak; interpretation of some observations. Tellus, 39, 245–253.
Sako, Y., Ishida, Y., Nishijima, T. and Hata, Y. (1987) Sexual reproduction and cyst formation in the freshwater dinoflagellate Peridinium penardii. Nippon Suisan Gakkaishi (Bull. Jpn Soc. Sci. Fish.), 53, 473–478.
Takayama, H., Matsuoka, K. and Fukuyo, Y. (1998) A taxonomic study on Gyrodinium aureolum Hulburt (Dinophyceae) from morphological viewpoint based on materials collected in Japanese coastal waters. Bull. Plankton Soc. Jpn, 45, 9–19.
Taylor, F. J. R., Fukuyo, Y. and Larsen, J. (1995) Taxonomy of harmful dinoflagellates. In Hallegraeff, G. M., Anderson, D. M. and Cembella, A. D. (eds), Manual on Harmful Microalgae. IOC Manuals and Guides No. 33. UNESCO, Paris, pp. 283–317.
Utermöhl, H. (1958) Zur Vervollkommnung der quantitativen phytoplanktonmethodik. Mitt. Int. Ver. Limnol., 9, 1–38.
Villafãne, V., Helbling, E., Holm-Hansen, O. and Chalker, B. (1995) Acclimatization of Antarctic natural phytoplankton assemblages when exposed to solar ultraviolet radiation. J. Plankton Res., 17, 2295–2306.
von Stosch, H. A. (1973) Observation on the vegetative reproduction and sexual life cycles of two freshwater dinoflagellates, Gymnodinium pseudopalustre Schiller and Woloszynskia apiculata sp. nov. Br. Phycol. J., 8, 105–134.
Wall, D. (1973) Modern dinoflagellate as a standard for paleontologic inquiry. In Evitt, R. E. (ed.), Proceedings of a Forum on Dinoflagellates. Anaheim, California, Contrib. Ser. Am. Ass. Stratigraphic Palynologists, pp. 37–43.
Water Quality Association of the Bohus Coast (1990–99) Kustvattenkontrollen i Göteborgs och Bohuslän. Årssammanfattningar 1990–99. County Administration of Göteborg and Bohuslän, Uddevalla (in Swedish).
Willén, T. (1962) Studies on the phytoplankton of some lakes connected with or recently isolated from the Baltic. Oikos, 13, 169–199.
Zohary, T., Pollingher, U., Hadas, O. and Hambright, K. (1998) Bloom dynamics and sedimentation of Peridinium gatuense in Lake Kinneret. Limnol. Oceanogr., 43, 175–186.
Received on April 5, 2000
; accepted on April 2, 2001
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Miyaguchi, T. Fujiki, T. Kikuchi, V. S. Kuwahara, and T. Toda Relationship between the bloom of Noctiluca scintillans and environmental factors in the coastal waters of Sagami Bay, Japan J. Plankton Res., March 1, 2006; 28(3): 313 - 324. [Abstract] [Full Text] [PDF]
|
|