Factors affecting outbreaks of high-density Cochlodinium polykrikoides red tides in the coastal seawaters around Yeosu and Tongyeong, Korea

Marine Pollution Bulletin 52 (2006) 1249–1259 www.elsevier.com/locate/marpolbul

Factors affecting outbreaks of high-density Cochlodinium polykrikoides red tides in the coastal seawaters around Yeosu and Tongyeong, Korea Young Sik Lee

*

South Sea Fisheries Research Institute, Division of Water Environment, 347 Anpo Hayang, Yosu Jeonnam 556-820, South Korea

Abstract Red tides caused by the dinoflagellate Cochlodinium polykrikoides occur annually in coastal waters of Korea, causing significant damage. A distinguishing characteristic of C. polykrikoides red tides is that they develop and persist in the open sea, where the water is comparatively clean with little contamination from the shore. We examined the causes of and key nutrients involved in high-density C. polykrikoides red tide outbreaks in the coastal seawaters around Yeosu and Tongyeong, Korea. High-density C. polykrikoides red tides occur in the coastal areas of Geomo Island, where freshwater flows into the sea after heavy rainfall events. Red tides are widespread in years when rainfall is heavy. The maximum concentration of C. polykrikoides and the duration of the red tides increase with increasing rainfall. Adding nitrogen and Seomjin River water to cultures of C. polykrikoides also increases biomass production and cell density of C. polykrikoides remarkably increased after heavy rainfall events. The occurrence of high concentrations of C. polykrikoides along the shores of Yeosu and Tongyeong seems to result from rainfallinitiated inflows of high concentrations of nitrate secondarily, after a conducive physical and chemical open-water environment has been established for C. polykrikoides to spread initially.  2006 Elsevier Ltd. All rights reserved. Keywords: Bioassay; Coastal area; Cochlodinium polykrikoides; Nitrate; Red tide; Rainfall

1. Introduction A red tide caused by Cochlodinium polykrikoides occurred for the first time in 1982 in the Nakdong River estuary and on the eastern side of Gaduck Island, Korea (Kim, 1997). Since that time, red tides have occurred annually in the coastal seawaters around Yeosu and Tongyeong, Korea. Recently, a red tide occurred for the first time off the coasts of Naro and Namhae Islands, and the occurrence of red tides is spreading throughout other coastal regions, including the East Sea (Sea of Japan) and off the Gunsan coast in the Yellow Sea. Red tides caused by C. polykrikoides result in significant damage to the marine

*

Tel.: +82 61 690 8958; fax: +82 61 686 1588. E-mail address: [email protected]

0025-326X/$ - see front matter  2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.marpolbul.2006.02.024

product industry, and prevention efforts are very expensive (Fig. 1, Table 1). Generally, harmful algal blooms caused by diatoms, such as Skeletonema costatum, Chaetoceros spp., Thalassiosira spp., and other species, develop in locations of large, temporary influxes of freshwater (Lee, 2002). In contrast, red tides caused by Heterosigma akashiwo and Prorocentrum spp., occur mainly in polluted inland seas or inner bays in Korea (Lee et al., 2002), such as off the northern coasts of Masan and Gamak Bays, Korea. Thus, the inflow of excess nitrogen and phosphorus from land appears to be the central cause of eutrophication and red tide development in coastal areas. Accordingly, prevention efforts have focused on reducing the amounts of nitrogen and phosphorus inflow. However, red tides caused by C. polykrikoides initially develop and tend to persist in the comparatively clean waters of outer bays and offshore areas.

1250

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

Fig. 1. Observed areas of C. polykrikoides red tides in coastal waters of Korea from 1994 to 2001 (data from http://www.nfrdi.re.kr).

48 84

55

3.2

34

1.6 15 21 746

29 29 54

East coast of Oenaro Island

2.6

57 42

August 2 (June 24) East coast of Naro and Dolsan Island August 14 (June 25) East coast of Naro and Dolsan Island

August 21 (July 22) East coast of Naro and Dolsan Island, South coast of Namhae Island 29 Auguts 9 (June 28)

August 29 (July 8) South coast of Oenaro Island August 25 (July 23) Coast of Naro Island, South coast of Namhae Island

Duration of the red tide (day) Fisheries damage (hundred million won)

The first observation time (lunar calendar) The first observation area

1251

C. polykrikoides is a phytoplankton that requires nutrients including nitrogen and phosphorus to grow in high concentrations. Nitrogen and phosphorus are the limiting nutrients in estuaries and inland sea ecosystems (Caraco et al., 1987; Harrison et al., 1990; Howarth, 1988). Thus, it should be possible to control occurrences of C. polykrikoides red tides by reducing the supply of either nitrogen or phosphorus. However, because C. polykrikoides red tides develop and persist in relatively clean coastal waters with almost no inflow of contaminants from land (Fig. 1, Table 1), it is of interest to determine how and why C. polykrikoides red tides occur. A recent study has related the occurrence of C. polykrikoides red tides to offshore seawater and submarine groundwater (Yang et al., 2000). Thus, questions about the current prevention measures for reducing nitrogen and phosphorus inputs from inland areas must be raised. Questions about the mechanism of C. polykrikoides redtide formation can be classified largely as follows. (1) Does C. polykrikoides form cysts, and, if so, when do the cysts germinate? (2) Why do red tides occur at the same time (around the neap tide) and in the same area (coasts of Naro and Namhae Islands) every year (Table 1) at low concentrations (<6000 cells/ml)? (3) Why do C. polykrikoides red tides develop at high concentrations? In this study, the causes of C. polykrikoides red tides at high concentrations and the key nutrients that cause this effect were examined in the coastal seawater around Yeosu and Tongyeong, Korea. 2. Experimental procedure

August 2 (June 25) Coast of Geomo Island

August 29 (August 4) Coast of Naro Island

September 4 (July 22) Coast of Naro and Geomo Island

1999 1998 1997 1996 1995 1994

Table 1 Outbreak characteristics of C. polykrikoides red tide in coastal waters of Korea from 1994 to 2002 (Lee et al., 2001; http://www.nfrdi.re.kr)

2000

2001

2002

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

2.1. Description of the study sites Naro, Geomo, Namhae, and Saryang Islands are centrally located in the south of Korea. In this region, the inflow of open seawater is affected by the Tsushima Current, which is a tributary of the Kuroshio Current. The city of Suncheon (population 267,000) is located north of Naro Island in the upper stream of Yeoja Bay. The Dong River (drainage area, 424 km2) flows through the center of Suncheon and into Yeoja Bay and the eastern coastal area off Naro Island. Freshwater from the 212.3-km long Seomjin River (drainage area, 4896 km2) flows into the northeast section of Gwangyang Bay. The cities of Gwangyang (population 136,000) and Suncheon are located near the bay. Industrial facilities, such as the Gwangyang steel mill, lie north of Gwangyang Bay, and Yeocheon National Industrial Development is in the south. Domestic sewage and wastewater from the factories flow into Gwangyang Bay. During periods of heavy rainfall, a large quantity of freshwater flows into the eastern coastal areas of Dolsan and Geomo Islands, while some of it flows into the southern coastal area of Namhae Island, passing the eastern coast of Yeosu (Lee, 2002). Jinyang Lake (total reservoir, 306 · 106m3) is in upper Jinju Bay. If freshwater inflows exceed its flood capacity,

1252

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

20% are discharged into the Nam River and 80% into Sacheon. Most of the freshwater discharged into Sacheon flows into the coastal area of Tongyeong’s Saryang Island through Jinju Bay (http://sacheon.kowaco.or.kr). 2.2. Sampling and analysis We intensively investigated at the eastern coast of Geomo Island after heavy rainfall events on August 2002 and August–October 2003 to seeing the effect of freshwater on red tide of C. polykrikoides (Fig. 2, Station AA). We sampled surface seawater to a depth of 0.5 m. Salinity was measured using an Environment Monitoring System (YSI, 6920). Dissolved inorganic nitrogen (DIN; NH4– N + NO3–N + NO2–N) and dissolved inorganic phosphorus (DIP; PO4–P) were measured after filtration with a Whatman GF/C filter using standard methods (Ministry of Maritime Affairs and Fisheries of Korea, 1998). The cell density of C. polykrikoides was determined by observing collected samples with an optical microscope (Olympus, BX 50). Data from C. polykrikoides red tides occurring between 1996 and 2003 in the coastal areas of Yeosu and Tongyeong were used in this study. The geographic ranges of the red tides were observed by eye and a portable microscope (Swift, FM-31). The classification and density of red tide organisms were determined by observing collected samples with an optical microscope (Olympus, BX 50). For the coastal area of Yeosu, we used precipitation data from the Yeosu and Suncheon Meteorological Station, since the freshwater sources are the Dong and Seomjin Rivers in Yeosu and Suncheon (http://www.kma.go.kr). For the coastal area of Tongyeong, precipitation data from the Jinju Meteorological Station was used, since most of the freshwater inflow is from the Jinju region (http:// www.kma.go.kr).

The average daily rainfall was calculated using the daily rainfall from 15 days before the first observation of a C. polykrikoides red tide until 2 days before the maximum concentration of C. polykrikoides was reached. We used a wide time range because laboratory incubations demonstrated that at certain times, seawater from coastal Yeosu and Tongyeong is conducive to dramatic increases in the growth of C. polykrikoides. At other times, it supports little or no increase in cell density of C. polykrikoides. These water conditions are related to the first occurrence and low levels of C. polykrikoides red tide, respectively (unpublished data). Therefore, we surmised that C. polykrikoides cannot develop to red tide density without certain physical and chemical factors present in the open-water environment. Even if the volume of freshwater flow into the coastal growth area is large, it may be difficult for C. polykrikoides to grow to high densities. Thus, the first observation time of a C. polykrikoides red tide is important as an indicator of the inflow of seawater conducive to the growth of C. polykrikoides. A previous study showed that salinity and nutrient concentrations returned to pre-rainfall levels 15 days after heavy rainfall, which led to low salinity (<17 psu) and high nutrient levels (Lee, 2002). Salinity dramatically decreased in the coastal area of Geomo Island, Yeosu, 2 days after heavy rainfall events at Suncheon (Lee, 2002). Statistical analyses were performed using the SPSS Windows Program (10.1, SPSS, Inc.). 2.3. Effects on biomass production by nitrogen, phosphorus, and Seomjin River water addition An algal assay was conducted with Seomjin River water to estimate the effect of freshwater on the growth of C. polykrikoides. The stock culture was maintained with Na2SiO3-excluded f/2 medium with seawater from the C.

Nam River

127-26N

128-10N Sacheon

Seomjin River Suncheon

34-28E

Goseong

Gwangyang Bay

Dong River

Yeoja Bay

Jinju Bay

Gwangyang

34-54E

Goheung

Jinju

Yeosu

Tongyeong

Namhae Island

Saryang Island

Gamak Bay Dolsan Island Yokji Island Geomo Island 0

15

30Km

Korea Study Area

Naro Island

Fig. 2. Map of the sampling stations to determine the spatial distributions of salinity, nutrients and C. polykrikoides concentration (AA) following rainfall events on the coast of Yeosu, and intensively studied areas of C. polykrikoides red tides along the coasts of Yeosu and Tongyeong, Korea.

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

polykrikoides growth region. (As mentioned above, C. polykrikoides could not be successfully cultured with other coastal seawaters from Yeosu and Tongyeong, Korea.) For the algal assay, 5 ml of surface seawater from Geomo Island were inserted into a 15-ml test tube after filtration with a 0.45-lm pore size filter, and C. polykrikoides were inoculated into the tube. This surface seawater was collected on 15 August 2003, just before a red tide developed, and the salinity, DIN, and DIP concentration were 32.41 psu, 1.02 lM and 0.13 lM, respectively. Concentrated stock solutions of NaNO3 and NaH2PO4 were added for nitrogen (882.35 lM) and phosphorus (36.232 lM) spikes at concentrations of f/2 medium. All compounds of Na2SiO3-excluded f/2 medium were spiked to confirm the growth of C. polykrikoides. No nutrients were added to the controls. Seomjin River water was added to the surface seawater at concentrations of 10% and 20%. The river water was filtered through a membrane filter of 0.45-lm pore size. DIN and DIP concentrations were 74.89 lM and 0.10 lM, respectively. The initial concentration of the inoculated C. polykrikoides was 200 cells/ml. Cultures were maintained at 23 ± 2 C, 140 ± 10 lmol m 2 s 1 brightness, and a 12:12-h cycle of light and darkness. The growth monitoring was conducted with a PHYTO-PAM chl fluorometer (Schreiber et al., 2002). 3. Results 3.1. Areas and ranges of high-density C. polykrikoides red tides Coastal areas around Yeosu and Tongyeong where dense (P6000 cells/ml) C. polykrikoides red tides occurred between 1996 and 2003 are shown in Figs. 3 and 4. Around Yeosu (Fig. 3), dense C. polykrikoides red tides did not occur along the northern coasts of Yeoja and Gamak Bays, which receive large inflows of freshwater and domestic sewage. High-density C. polykrikoides red tides also did not occur in the open sea, where freshwater inflow is comparatively low (south coast of Naro and Geomo Islands). Dense red tides mainly occurred along the east coasts of Naro and Dolsan Islands and off the coasts of Geomo Island. In 1997, when average rainfall was relatively low (average daily rainfall 3.0 mm/day), red tides in densities exceeding 6000 cells/ml occurred only along sections of the east coast of Naro Island. In 1999 (average daily rainfall 14.8 mm/ day) and 2002 (average daily rainfall 18.6 mm/day), when rainfall was high, C. polykrikoides red tides covering large areas occurred along the east coasts of Naro and Geomo Islands. There was no direct correlation between average daily rainfall and C. polykrikoides red tide areas in 2001. In the Tongyeong area (Fig. 4), dense C. polykrikoides red tides were not observed in Jinju Bay, which is affected by domestic sewage and freshwater, nor along the southern coast of Yokji Island, which faces open seawater. Dense C. polykrikoides red tides occurred mainly along the coastal

1253

region of Saryang Island and the western coast of Tongyeong, where the effects of domestic sewage and freshwater are less than in Jinju Bay, where the effect of open seawater is less than at the south coast of Yokji Island. As was the case for the coastal area of Yeosu, dense C. polykrikoides red tides occurred more widely during the wet years of 1999 and 2002 than in 1998, when average rainfall was lower. 3.2. Correlation between salinity and C. polykrikoides concentration The correlation between salinity and a C. polykrikoides concentration of >6000 cells/ml in the coastal areas of Yeosu from 1996 to 2003 is shown in Fig. 5. At salinities of 32 psu or higher, fewer than 7000 cells/ml of C. polykrikoides were measured. However, at salinities of less than 29 psu, more than 9000 cells/ml of C. polykrikoides were observed. Thus, lower salinity was correlated with a high concentration of C. polykrikoides, whereas higher salinity was correlated with a lower concentration of C. polykrikoides (r2 = 0.313, p = 0.000). 3.3. Average daily rainfall and highest concentration of C. polykrikoides red tide The correlation between the highest annual concentration of C. polykrikoides and the average daily rainfall on the coasts of Yeosu and Tongyeong from 1996 to 2003 is shown in Fig. 6. Along the coast of Yeosu, the average daily rainfall was 3.0 mm in 1997, and the highest concentration of C. polykrikoides was 7400 cells/ml. In 1999, an average daily rainfall of 14.8 mm was recorded, and the highest concentration of C. polykrikoides was 35,000 cells/ml. Thus, as the average daily rainfall increased, the concentration of C. polykrikoides also tended to increase. In 1998, the coastal areas of Tongyeong received almost no rainfall (0.2 mm/day on average), and the concentration of C. polykrikoides (7000 cells/ml) was the lowest of the years investigated. In contrast, the coast of Yeosu received 7.1 mm of average daily rainfall that year, and the highest concentration for that area was 25,000 cells/ml. In 1999, the average daily rainfall in Tongyeong was 16.0 mm, and the highest concentration of C. polykrikoides was 25,000 cells/ml. Here, as at Yeosu, the trend was for the concentration of C. polykrikoides to increase as average daily rainfall increased. 3.4. Average daily rainfall and the duration of dense C. polykrikoides red tides The number of days that C. polykrikoides exceeded 6000 cells/ml plotted against the average daily rainfall in the coastal areas of Yeosu and Tongyeong from 1996 to 2003 is shown in Fig. 7. For the Yeosu coast, 3.0 mm of average daily rainfall was recorded in 1997, and the two days when more than 6000 cells/ml of C. polykrikoides

1254

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

Fig. 3. Annual distributions of high density (>6000 cells/ml) C. polykrikoides red tides in the coastal areas of Yeosu, Korea from 1996 to 2003. Numbers in each figure are maximum concentrations of C. polykrikoides.

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

1255

Fig. 4. Annual distributions of high density (>6000 cells/ml) C. polykrikoides red tide in the coastal areas of Tongyeong, Korea from 1996 to 2003. Numbers in each figure are maximum concentrations of C. polykrikoides.

red tide were counted was the lowest number of days in the study. As average daily rainfall increased, the duration of

red tides also increased, with the exception of 2000. In 2002, when rainfall was the heaviest, the red tide lasted

1256

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259 12

R2=0.313, p=0.000, n=47

Coastal waters around Yeosu

10

4.4

Number of outbreak days

log cell concentration (cells/mL)

4.6

4.2

4.0

3.8

1999 2003

8

2002

6 1998

4

2001

1996 2000

2

2

1997

R =0.61, p=0.023

0

3.6 26

27

28

29

30

31

32

2

33

4

6

Salinity (psu) Fig. 5. Relationship between salinity and maximum concentration of C. polykrikoides in the coastal areas of Yeosu between 1996 and 2003. Data are for concentrations of C. polykrikoides >6000 cells/mL.

10

12

14

16

18

20

Coastal waters around Yeosu 1999

30000

2003

14 12

Number of outbreak days

40000

1998

8

Average rainfall (mm/day)

2002

20000

Coastal waters around Tongyeong

1997 2002

10 2003

1999

8 2000 1996

6 4

2001 2

2

R =0.62, p=0.021

1998

0 0

2001

1996

10000

2

4

6

8

10

12

14

16

18

Average rainfall (mm/day) 2000

1997

2 R =0.43, p=0.079

0 0

5

10

15

20

Average rainfall (mm/day)

recorded. As the amount of rainfall increased, the duration of red tides also increased proportionately to reach 12 days in 2002. Thus, both Tongyeong and Yeosu showed the same trend of increased rainfall correlating with increased duration of red tides.

30000 Coastal waters around Tongyeong

25000

2003 2001

1999

20000

3.5. Biomass production by the addition of nitrogen, phosphorus, and Seomjin River water

2002

1997

15000 1996

2000

10000 1998

5000

2

R =0.21, p=0.249

0 0

2

4

6

8

10

12

14

16

Fig. 7. Relationship between average daily rainfall and number of outbreak day (>6000 cells/ml) in each year in coastal waters around Yeosu and Tongyeong. Numbers in the figures represent years.

18

Average rainfall (mm/day)

Fig. 6. Relationship between average daily rainfall and maximum concentration of C. polykrikoides in each year in coastal waters around Yeosu and Tongyeong. Numbers in the figures represent years.

for 9 days. At Tongyeong, the number of days when C. polykrikoides exceeded 6000 cells/ml was lowest in 1998 (2 days), when the least amount of rainfall was also

When nitrogen and f/2 medium were added to surface seawater from Geomo Island, mean chlorophyll a values was increased to more than 13 times that of the control (no nutrient addition; Fig. 8). On the other hand, the addition of phosphorus alone had no effect on C. polykrikoides biomass accumulation. The addition of 10% and 20% Seomjin River water increased chlorophyll a values by 6 and 12 times, respectively, relative to control values. 3.6. Temporal and sequential study after heavy rainfall events Fig. 9 shows the temporal variations of rainfall, salinity, DIN, DIP, and cell density of C. polykrikoides after heavy rainfall events at the eastern coast of Geomo Island

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

Concentration of Chl.a (µg/L)….

160 140 120 100 80 60 40 20 0 C

N

P

F/2

10% RW 20% RW

Fig. 8. Effects on C. polykrikoides biomass production (maximum PhytoPam chlorophyll a values) incubated without additions (C = control, no additions), and additions of nitrate (N), phosphate (P), f/2 medium elements (F/2), Seomjin River water (10% RW = added to give 10% vol/ vol Seomjin River water, 20% RW = added to give 20% vol/vol Seomjin River water) using good seawater for growth of C. polykrikoides. All treatments were triplicates (n = 3). Error bars indicate standard deviations (±SD).

in August, 2002 and August–October, 2003. Salinity in August, 2002 and August–October, 2003 varied from 26.5–33.0 and 24.10–32.00 psu, respectively. Salinity was high on 2–5 August, 2002 and 12–18 August, 2003. The

1257

salinity decreased until 26.5 psu on 19 August, 2002, 24.1 psu on 22 August, and 14 September, 2003 and the drops were associated with rainfall events. DIN concentration in August 2002 and August–October, 2003 varied from 0.51–29.45 lM and 0.34–34.88 lM, respectively. High DIN concentration was observed with low salinity. DIP concentration in August, 2002 and August–October, 2003 varied from 0.09–1.35 lM and 0.11–1.50 lM, respectively. Cell density of C. polykrikoides was 270–950 cells/ml on August 2–5, 2002 and 120–350 cells/ml on August 12– 14, 2003 before heavy rainfall events. After that or/and low salinity observed, the cell density was remarkably increased on August 10 (12,750 cells/ml), August 20, 2002 (25,000 cells/ml), August 29 (9500 cells/ml), and September 26, 2003 (26,100 cells/ml). No C. polykrikoides red tide was observed after September 1, 2002 in spite of heavy rainfall events due to typhoon Rusa on August 31, 2002. 4. Discussion In this study, high-density C. polykrikoides red tides occurred in locations where freshwater entered after heavy rainfall events (Figs. 3, 4, 9). Moreover, the geographical range of the red tides was broad in the years when there was increased rainfall (Figs. 3, 4) and decreased salinity (Fig. 5). In addition, as the amount of rainfall increased,

3

30

2

20

1

10

0

0

34

50

32 30

30000 25000 Typhoon Rusa

100 150

15000

200

10000

26

250

5000

24

300

28

0

1

10

20

30

40

3

30

2

20

1

10

0

0

36

0

34

50

Rainfall (mm/day)

4

Salinity (psu)

50

DIN(uM)

DIP(uM)

August 5

20000

Concentration of Cochlodinium poykrikoides (cells/mL)

40

0

Concentration of Cochlodinium poykrikoides Salinity DIP DIN

10

September 2002 30000 25000

Typhoon Maemi

100

20000

150

15000

200

10000

26

250

5000

24

300

32 30 28

Concentration of Cochlodinium poykrikoides (cells/mL)

4

36

Rainfall (mm/day)

50

Salinity (psu)

5

DIN(uM)

DIP(uM)

Rainfall (mm/day)

0 1

August

10

20

30

10

September

20

30

10

October 2003

Fig. 9. Temporal variations of rainfall, salinity (psu), DIN (lM), DIP (lM), and concentration of C. polykrikoides after heavy rainfall events at the eastern coast of Geomo Island (st. AA) on August 2002 and August–October 2003.

1258

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259

the maximum concentration and duration of the red tides also increased (Figs. 6 and 7). Finally, C. polykrikoides biomass production increased when nitrogen and Seomjin River water were added to cultures (Fig. 8), and cell density of C. polykrikoides remarkably increased after heavy rainfall events (Fig. 9). Equally crucial to the development of a C. polykrikoides red tide is the inflow of seawater conducive to the growth of this species. For instance, despite a large quantity of freshwater entering the coastal seawaters of Geomo Island after September 1, 2002, no C. polykrikoides red tides were observed (Fig. 9). A bioassay was conducted using two surface seawater samples collected at the same time, one from the east coast of Geomo Island and the other from the west coast of Geomo Island. The biomass of C. polykrikoides on the east coast increased remarkably, but the west coast showed little growth, despite the samples having comparable salinity and nutrient levels (unpublished data). These data suggest that the development of high-density C. polykrikoides red tides in the coastal areas of Yeosu and Tongyeong seems to be triggered by the inflow of freshwater from rainfall events after the input of seawater conducive to the growth of C. polykrikoides. Kim (2003) summarized the 23-year datasets of precipitations and occurrence of C. polykrikoides red tide in Yatsushiro Sea, Japan. The result revealed that large-scale C. polykrikoides red tide has occurred in low precipitations season and declined the red tide by heavy rainfall events. In this study, no C. polykrikoides red tide was also occurred after heavy rainfall event on September 1, 2002 (Fig. 9). As mentioned above, there is good seawater for growth of C. polykrikoides except for nutrients due to freshwater. Therefore, the reason why do not red tides occurred after heavy rainfall event in September, 2002 seems to be due to not input of seawater conducive to the growth of C. polykrikoides. This means that C. polykrikoides red tide was not occurs at all times after heavy rainfall event. In this study, high-density C. polykrikoides red tides (>6000 cells/ml) occurred at >27 psu of salinity and the minimum salinity was 24.10 psu at the eastern coast of Geomo Island (Figs. 5 and 9). According to Kim et al. (2004) paper, the maximum growth rate of C. polykrikoides was obtained at 25 C and 34 psu of salinity and the growth rate was decreased with low salinity (Kim, 2003). However, the paper showed that C. polykrikoides was propagated at 24 psu with >0.2 day 1 of growth rate (Kim et al., 2004). Therefore, 24.10 psu of salinity will not limit to C. polykrikoides growth. Large quantities of nitrogen and phosphorus, as well as other trace elements, are required to develop high-density C. polykrikoides red tides. In this study, we observed no notable fluctuations in the concentrations of DIP (Fig. 9). Additionally, the overall concentration of inorganic phosphorus was comparatively lowered than that of inorganic nitrogen in freshwater. In freshwater, the limiting nutrient for phytoplankton growth is phosphorus (Harrison et al.,

1990). In the case of DIN in this study, a large quantity of nitrate was introduced by freshwater inflows into the coastal areas where C. polykrikoides red tides occurred (Fig. 9). Nitrogen is the usual limiting nutrient for phytoplankton growth in the coastal seawater of Geomo Island (Lee, 2002). In our cultures, the biomass of C. polykrikoides was increased by the addition of nitrogen (Fig. 8). Accordingly, we can conclude that the occurrence of high-density C. polykrikoides red tides in the coastal areas of Yeosu and Tongyeong is facilitated by the optimal physical (e.g., light) and chemical (trigger element(s)) environmental conditions, and is then promoted by the introduction of a large quantity of nitrate following heavy rainfall. 5. Conclusions In this study, we examined the physical and nutritional causes of high-density C. polykrikoides red tides in the coastal seawaters around Yeosu and Tongyeong, Korea. Specific conclusions derived from this study are as follows: (1) High-density C. polykrikoides red tides occur in the coastal areas of Geomo Island, where freshwater flows into the sea after heavy rainfall events. (2) As average daily rainfall increases, the maximum concentration of C. polykrikoides and the duration of red tides also increase. The addition of nitrogen and Seomjin River water to cultures of C. polykrikoides also augmented biomass production and cell density of C. polykrikoides remarkably increased after heavy rainfall events. (3) Occurrences of high-density C. polykrikoides red tides along the seashores of Yeosu and Tongyeong seem to result from rainfall-initiated inflows of high levels of nitrate after a suitable physical and chemical openwater environment has been created for C. polykrikoides to spread initially. Acknowledgment This work is funded by a grant from the National Fisheries Research and Development Institute (RP-05-ME-1). References Caraco, N., Tamse, A., Boutros, O., Valiela, I., 1987. Nutrient limitation of phytoplankton growth in brackish coastal ponds. Can. J. Fish. Aquat. Sci. 44, 473–476. Harrison, P.J., Hu, M.H., Yang, Y.P., Lu, X., 1990. Phosphate limitation in estuarine and coastal waters of China. J. Exp. Mar. Biol. Ecol. 140, 79–87. Howarth, R.W., 1988. Nutrient limitation of net primary production in marine ecosystems. Ann. Rev. Ecol. 19, 89–110. Kim, H.G., 1997. Harmful algal blooms in Korean coastal waters, with a focus on three fish-killing dinoflagellates. In: Proceedings of the Korea–China Joint Symposium on Harmful Algal Blooms hosted by the National Fisheries Research and Development Institute and the Chinese Academy of Fisheries Science, Pusan, Korea, 5–7 December, pp. 1–20.

Y.S. Lee / Marine Pollution Bulletin 52 (2006) 1249–1259 Kim, D.-I., 2003. Physiological and ecological studies on harmful red tide dinoflagellate Cochlodinium polykrikoides(margalef). Doctoral thesis, Kyushu University, p. 154. Kim, D.-I., Matsuyama, Y., Nagasoe, S., Yamaguchi, M., Yoon, Y.-H., Oshima, Y., Imada, N., Honjo, T., 2004. Effects of temperature, salinity and irradiance on the growth of the harmful red tide dinoflagellate Cochlodinium polykrikoides Margalef (Dinophyceae). J. Plankton Res. 26, 61–66. Lee, Y.S., 2002. The influence of inflowing freshwater on diatom blooms along the eastern coast of Dolsan, Yosu, Korea. J. KSEE 24, 477–488. Lee, Y.S., Park, Y.T., Go, W.-J., Kim, K.Y., Park, J., Jo, Y.-J., Park, S.Y., 2001. Countermeasure and outbreak mechanism of Cochlodinium polykrikoides red 1. Environmental characteristics on outbreak and disappearance of C. polykrikoides bloom. J. Korean Soc. Oceanogr. 6, 259–264.

1259

Lee, S.G., Kim, H.G., Bae, H.M., Kang, Y.S., Jeong, C.S., Lee, C.K., Kim, S.Y., Kim, C.S., Lim, W.-A., Cho, U.S., 2002. Handbook of Harmful Marine Algal Blooms in Korean Waters. Nat. Fish. Res. Devel. Inst., Republic of Korea, p. 172. Ministry of Maritime Affairs and Fisheries of Korea, 1998. Method of seawater analysis. Schreiber, U., Gademann, R., Bird, P., Ralph, P.J., Larkum, A.W.D., Kuhl, M., 2002. Apparent light requirement for activation of photosynthesis upon rehydration of desiccated beachrock microbial mats. J. Phycol. 38, 125–134. Yang, J.S., Choi, H.-Y., Jeong, H.J., Jeong, J.Y., Park, J.K., 2000. The outbreak of red tides in the coastal waters off Goheung, Chonnam, Korea: 1. Physical and chemical characteristics in 1997. J. Oceanol. Soc. Korea 5, 16–26.