An outdoor enclosure experiment was carried out near the southern coast of Korea in order to investigate the effect of enclosed environments on planktonic community structure. The experiment was conducted in 6 floating, 220-liter cylindrical enclosures maintained for short periods (17 days). Rapid decreases of inorganic nutrients (N. P, Si), and in the abundance of large diatoms, Coscinodiscus spp., were observed after only one day from the start of the experiment. Decreases in chlorophyll a progressed through the experimental period, although the small diatoms, Chaetoceros spp. and Thalassiosira sp., increased considerably during the initial days. The continuous supply of dissolved organic carbon from crashing stocks of planktonic organisms sustained rapid growth of heterotrophic bacteria. The bacterial increase was followed by increase of heterotrophic nanoflagellates and ciliated protozoa. Noctiluca scintillans flourished as a dominant zooplankter until the end of the experiment, feeding on phytoplankton during the first half of the observations and on the abundant heterotrophic microorganisms during the second half. The results suggested that the enclosed environment provoked a structural shift of the plankton community from autotrophy to heterotrophy and indicated additionally that the trophic status in plankton communities was reflected by an enclosed environment formed by artificial structures in the coastal zone.
Natural barriers enclosing embayments and coastal engineering structures, such as breakwaters, reduce water circulation from the oligotrophic ocean, causing eutrophication. Chronic eutrophication by nutrients steadily received from inflowing waters leads to blooms of heterotropic microorganisms in the sheltered areas, as well as high chlorophyll a (chl a). Changes in planktonic microbial communities of coastal environments enclosed by breakwaters have been reported for two coastal areas of Korea. The enclosed environments develop abundant communities of planktonic microbes comprised of heterotrophic bacteria, heterotrophic nanoflagellates and ciliates (Kim et al., 2007a, 2007b). However, despite the clear distinctions between the community structure in enclosed areas and that offshore, it cannot simply be concluded that the biological response results from the physical barriers enclosing such areas. Controlled experiments are required to demonstrate the effects of environmental enclosure on planktonic communities because there are many potentially confounding physical and biological factors in the field.
Experimental ecosystems are a useful tool for investigating the effects of enclosure on a plankton ecosystem (Jung et al., 2010). Studies involving enclosures have concentrated on energy and material transfer from one trophic level to another, and on interactions among nutrients, phytoplankton and zooplankton. The size and shape of an enclosure can have important effects on the ecological dynamics inside (Petersen et al., 2009). Therefore, these effects need to be understood and taken into account in designing and interpreting observations in enclosures. It is essential to understand the effects of enclosure itself prior to comparisons of different treatments in marine experimental ecosystems.
To reach that understanding, we set up an outdoor enclosure experiment to investigate the environmental characteristics and the changes of planktonic community structure resulting from the design characteristics of enclosures. Our observations were concentrated on the dynamics of the microbial food web in the enclosures. Heterotrophic communities in the enclosed pelagic ecosystem presented four functional trophic components: bacteria, nanoflagellates, ciliates, and Noctiluca.
Materials and Methods
An enclosure experiment was carried out at the Jangmok marine station (34° 59′ 38″ N, 128° 40′ 28″ E) located on Geoje Island in the southern coastal waters of Korea, starting on 30 September 2008. The experimental set-up consisted of three transparent, cylindrical plastic enclosures 150 cm in height and 50 cm in diameter, which were filled with 220-liter of coastal water collected from the surface layer at the marine station. The enclosures were deployed in the upper waters (0 to 1.0 m depth) of the marine station. The whole water column was stirred vertically using a rod for 5 min 3 times each day and completely mixed before the sampling. In situ water was held in the enclosures without any manipulation to investigate the enclosed effect in comparison with ambient natural waters. The enclosure observations were carried out in for 17 days with sampling from each enclosure once each day at 10:00 AM.
Water samples were collected from a depth of 0.5 m using Van Dorn bottles in the enclosures and the ambient waters alongside. Planktonic components, inorganic nutrients, chl a concentration, and dissolved organic carbon (DOC) were analyzed. Water temperature and practical salinity were recorded using a conductivity-temperature-depth recorder. Dissolved oxygen concentration and pH were measured using digital instruments. chl a, inorganic nutrients, and DOC were analyzed by a manual of chemical and biological methods for seawater analysis (Parsons et al., 1984).
Samples for bacteria and nanoflagellates were preserved immediately in glutaraldehyde and stained with 4′,6-diamidino-2-phenylindole (DAPI) for bacteria (Porter and Feig, 1980) and fluorescein isothiocyanate (FITC) for nanoflagellates (Sherr et al., 1993). The stained cells were counted via fluorescence microscopy. Subsamples of microplankton (diatoms, dinoflagellates, and ciliates) were fixed in Lugol's solution (final conc. v/v 2%). Microplankton cells were enumerated and identified under a microscope. For Noctiluca analysis, 8-liter subsamples from the water samples were filtered with a plankton net (mesh size 200 μm) and preserved in buffered formalin (final conc. 5%). Noctiluca was counted under a stereo microscope. Planktonic components were analyzed with the same methods in both the enclosure and the ambient water.
Environment and planktonic components were compared based on each average value of three enclosures and of three ambient waters. Significant differences of abiotic factors between the enclosed and ambient water were assessed by paired t-test using the SAS statistical program. Data on biotic and abiotic factors were evaluated in part by multivariate analyses using PRIMER 6: cluster analysis was performed using group average clustering from the similarity by a Bray-Curtis method. Using the ranked similarity matrix, an ordination plot was produced by non-metric multidimensional scaling (MDS).
Water temperature ranged from 22.2 to 24.0°C and showed no significant difference between enclosed and ambient water (Table 1). A slight increase of salinity from 32 to 33 psu was detected in the enclosed water over time due to evaporation. pH increased to 8.6 in ambient water during the latter half, while remaining relatively steady at 8.2–8.3 in the enclosure. There was no clear change in dissolved oxygen (DO) between enclosed and ambient water within the first 3 days, but then it increased to above 8.0 mg l−1in the ambient water. Concentrations of nitrogenous nutrients (DIN) and phosphate remained at very low levels, below 0.5 μmol after a day. The silicate concentration in the enclosure also showed a gradual decrease. In contrast, higher concentrations of inorganic nutrients were sustained during the first half of the experimental period in ambient water. There was a gradual increase of DOC within the first 4 days in the enclosure. In ambient water, DOC recorded an averaged concentration of 5.2 mg l−1 without a significant change. Thus, DOC did not differ significantly in both waters while inorganic nutrients were lower in the enclosure compared with the ambient water (Table 1).
Autotrophic planktonic components
In enclosure water, chlorophyll a gradually decreased over time to a low of 0.3 μg l−1. Chl a decrease in the enclosure concurred with the decline of inorganic nutrients. Changes in microphytoplankton abundances showed a clear difference between the enclosure and ambient water (Figure 1a). A rapid increase in the enclosure was observed during the first 3 days, generating the maximum abundance of ca. 2 × 106 cells l−1. Thereafter, the abundance continuously decreased until the end of experiment. In comparison, the abundance of microphytoplankton in ambient water slowly increased from the start of the experiment and reached a peak after 10 days. The dominant microphytoplankton species also differed between the two environments. Chaetoceros spp. and Coscinodiscus spp. were dominant in the enclosure. The abundance of Chaetoceros spp. displayed a pattern similar to the change of the total abundance of microphytoplankton. The abundance of Coscinodiscus spp. including the giant species Coscinodiscus wailesii, fell sharply after a day and then continued a gradual decrease (Figure 1c). In ambient water, Skeletonema costatum was the major contributor to the change in total abundance. The large dinoflagellate species Akashiwo sanguinea co-occurred during the increasing trend of S. costatum, but then A. sanguinea declined between days 10 and 12 in a manner that differed from the continuous growth of S. costatum (Figure 1d).
Heterotrophic planktonic components
Heterotrophic bacteria (HB) in the enclosures rapidly increased during the first two days to their highest abundance (8.2 × 107 cells ml−1) and maintained the higher level until the end of experiment period. In contrast to the enclosure, no significant change of HB was observed in the ambient water (Figure 2a). The abundance of heterotrophic nanoflagellates (HNF) also increased rapidly in the enclosure between day 2 and 3, and then remained at an average level of 104 cells ml−1, while a steady abundance was measured in ambient water (Figure 2b). Ciliate abundances in the enclosure showed an increase during the initial days, reaching a maximum of 6.5 × 103 cells l−1 (Figure 2c). A considerable difference of Noctiluca scintillans abundance was recorded between ambient and enclosure water, showing a rapid increase in the enclosure during the first half and peaked at 489 indiv. l−1 while just below 109 indiv. l−1 in ambient water. N. scintillans flourished until the end of experiment period in the enclosure (Figure 2d). N. scintillans abundance showed highly positive relationship with microphytoplankton abundance in the ambient water and a negative one in the enclosure (Figure 3).
The present study investigated the planktonic communities and related environments over a 17 day period in enclosures containing 220-liter seawater. There are three main findings in respect to the abiotic conditions: (1) there was good agreement of water temperature between the enclosure and nature; (2) salinity, pH and DO remained close to the initial conditions; (3) there was a rapid drop of inorganic nutrients, but only gradual change of dissolved organic carbon. As small organisms like plankton tend to have short generation times and small habitat ranges (Sheldon et al., 1972), experiments conducted in small experimental ecosystems and over short time periods are justifiable (Petersen et al., 2009). It is considered that the small enclosure can be applied as a practical tool for assessing short-term environmental effects on plankton communities.
Rapid drops of inorganic nutrient concentrations compared to the fluctuations of other abiotic parameters were the prominent events in the enclosed water. Increase in phytoplankton abundance during the initial three days was certainly based on the consumption of those nutrients. However, the increase in phytoplankton abundance even carbon biomass, (data not shown) was not strongly reflected by the increase in chlorophyll a. Dominant species generating the increase of phytoplankton abundance were small-sized species represented by Chaetoceros spp. and Thalassiosira sp. Nanophytoplankton has a competitive advantage by having faster nutrient uptake rates (Friebele et al., 1978). The rapid growth of small-sized species in the present enclosures may have been facilitated by their superior competition for limited nutrients. In contrast, the abundance of large diatoms like Coscinodiscus wailesii was largely reduced during the same period. C. wailesii has a higher content of chl. a (3.7–6.6 ng cell−1) than most other diatoms, and hence it is recognized as a very important contributor to the phytoplankton biomass (Tada et al., 2000). Thus, the decline of Coscinodiscus species was tightly related to a decrease in chl a, even though there was a significant increase of total phytoplankton numbers.
There are many reports of clear increases of DOC during or after phytoplankton blooms (Riemann et al., 2000). Dissolved organic carbon has mainly been considered to derive from autochthonous sources. The gradual increases of DOC in contrast to the rapid drop of inorganic nutrients in the present enclosure can also be understood to result from autochthonous production in the enclosure. It is well known that DOC serves as an energy source for heterotrophic microorganisms (Fenchel, 1982). Data from the present enclosure study suggest that the peak HB abundance that developed during the early days resulted from the abundant DOC (Figure 1b). Subsequent progress of the microbial loop resulted in a further increase of HB and led to the increase of predators that feed on HB. That is, HB abundances fell as the abundance of HNF and ciliates increased (Figure 2b, c). A similar result of bacterial number decrease has been found during the flagellate bloom in a seawater enclosure (Larsen et al., 2001). Interestingly, it is also commonly thought that ciliates act as competitors of HNF for HB prey. Therefore, DOC, HB, HNF, and ciliates are serially connected in the microbial food chain in the enclosure. This entire trophic sequence flourished up to the end of the 17-day manipulation.
A wide variety of prey items has been reported for Noctiluca scintillans, including bacteria, protozoans, and phytoplankton (Shanks and Walters, 1996). Microphytoplankton, heterotrophic bacteria, nanoflagellates and ciliates were serially available as food items for N. scintillans in the present enclosure. A negative relationship was found between Noctiluca and phytoplankton abundances in the enclosure, while a positive one was found in the ambient water (Figure 3). The negative relationship indicates that N. scintillans fed on HB and HNF as the major food sources and sustained its population well when the phytoplankton was depleted during the last half of the experimental period.
Figure 4 shows a cluster analysis and two-dimensional MDS ordination plot of samples based on environment factors measured and all planktonic components. A clear separation in MDS configuration indicated between ambient and enclosed water. A very clear difference in planktonic communities has been noted between inner and outer areas divided by an artificial breakwater in coastal waters (Kim et al., 2007a, 2007b). Heterotrophic bacteria, nanoflagellates, and ciliates develop higher biomass in the enclosed inner area. The results of the present experiment using seawater enclosures, in which heterotrophic plankton communities developed serially, while inorganic nutrient depletion weakened the autotrophic process, enhance our understanding of the structural shift of planktonic communities in enclosed coastal water systems. In addition, it can be recognized that the trophic status in plankton communities are reflected by enclosed environment formed by artificial structures in the coastal zone. There is a very large difference in scale between a breakwater enclosing an embayment and the present enclosures. An extrapolation from the outdoor enclosures to likely processes in small harbors will be conducted in a series of scaled-up mesocosms that differ in enclosure depth and radius.
To investigate variations in plankton communities on enclosed environments, the experiment was conducted in short-term marine mesocosms. In enclosed mesocosms, inorganic nutrients were decreased rapidly by consumption for growth of phytoplankton as energy sources. Continuous supplementation of dissolved organic carbon by degradation from dead planktonic organisms caused rapid growth of heterotrophic bacteria. Both heterotrophic nanoflagellates and ciliated protozoa were more abundant, which probably was based on bacteria as food source. Noctiluca scintillans flourished continually as a dominant zooplankter during experiment periods, feeding on phytoplankton and bacteria. The results suggested that the enclosed environment provoked a structural shift of the plankton communities from autotrophy to heterotrophy at artificial structures in the coastal zone.
We would sincerely like to thank Dr. C. B. Miller for valuable comments on the manuscript. This work was supported by the Pioneer Research Program for Converging Technology of the Ministry of Education, Science and Technology, Republic of Korea (2008-2000122).