NO. 4, 1982] OWEN AND NORRIS — GONYAULAX CYSTS 227

8. Total Natural Energies: Sun - 4.31 x 10 12 CE Cal/yr Chemical potential of rain - 2.76 x 10 12 CE Cal/yr Tidal energy - 1.04 x 10 12 CE Cal/yr Wind - 1.00 x 10 12 CE Cal/yr Kinetic energy of rain - 0.12 x 10 12 CE Cal/yr Waves - 0.16 x 10 12 CE Cal/yr Kinetic energy of river - 0.02 x 10 12 CE Cal/yr

9.41 x 10 12 CE Cal/yr

Biological Sciences

BENTHIC RESTING CYSTS OF GONYAULAX MONILATA HOWELL AND THEIR RELATIONSHIP TO RED TIDES IN THE INDIAN RIVER, FLORIDA

Kevin C. Owen and Dean R. Norris

Department of Oceanography and Ocean Engineering, Florida Institute of Technology, Melbourne, Florida 32901

Abstract: Gonyaulax monilata cysts were found in the sediments from 13 of 37 stations sampled in the Indian River, Florida. The distribution of cysts appears to be the result of 2 unrelated phenomena. In areas where blooms have occurred in the past, the lack of circulation in the lagoon allows the cysts to accumulate. South of Turkey Creek, cysts are found associated with shellfish beds where no blooms have been reported. It is suggested these cysts have been intro- duced into shellfish regions with oysters which are annually rafted from areas where G. monilata blooms occur. Blooms of this toxic dinoflagellate can be expected to reoccur in regions they have in the past, and the possibility of blooms in the vicinity of the shellfish beds around Grant, Florida now exists. However, there is not sufficient evidence to implicate direct toxicity to shellfish by G. monilata.

Hirano (1967) and Prakash (1967) independently suggested the possibil- ity that the similar year to year timing and location of dinoflagellate blooms

are related to benthic resting cysts (hypnocysts) . Several authors (e.g., Steidinger, 1973, 1975a, b; Wall, 1975; Dale, 1977; Anderson and Wall, 1978; Anderson and Morel, 1979a, b; Haddad and Carder, 1979; Anderson, 1980) have advanced the hypothesis that blooms might be initiated by ben- thic resting cysts. Steidinger (1975a) suggested that localized areas of ac- cumulation of benthic resting cysts act as seed "beds" and that such should be delimited in order to pinpoint origin of blooms.

The Indian River is a shallow lagoonal system on the east coast of Florida separated from the Atlantic Ocean by barrier islands. Periodic blooms in this lagoon are caused by the catenate dinoflagellate Gonyaulax monilata Howell (1953). Walker and Steidinger (1979) observed cyst formation in G. monilata under laboratory conditions in nitrogen-deficient medium and in stationary cultures. Similar cysts were first tentatively identified in Indian River samples and then conclusively identified by excystment of cysts from 228 FLORIDA SCIENTIST [Vol. 45

Tampa Bay sediments (Walker and Steidinger, 1979). Following initial iden- tification of G. monilata cysts from the Indian River, the sediments of the lagoon in south Brevard County were sampled to determine the distribution and density of this type of cyst. Our results indicate the cyst distribution is determined by the areas of past G. monilata blooms, the circulation patterns found in the lagoonal system, and the activity of commercial shell fishermen who regularly relay (transplant) oysters from areas closed to harvesting in the Banana and Indian rivers to lease areas in the Indian River. Methods — Between 3 February and 17 April, 1979 grab samples were taken at 37 stations in the Indian River between the Pineda Causeway and

Sebastian Inlet (Fig. 1). Fifteen stations are common to those periodically sampled by the Brevard County Health Department (B.C.H.D.); thus, some historical hydrographic and biological data are available. The remaining stations were selected for their proximity to tributaries, shellfish beds, or areas where discolored water has been observed in the past. Using a Ponar grab, 2 sediment samples were taken at each station. From the first, approximately 300 g of sediment to be used for grain size analysis were placed in a 470 ml jar. The second sample was placed in a 19 1 plastic

BANANA R.

8O>30' 28°10'

ATLANTIC

OCEAN

W

Fie. 1. Geographical distribution of Gonyaulax monilata benthic cysts, filled circles, in the Indian River, Florida. Open circles denote no G. monilata cysts found in sample. EC, Eau Gallie Causeway; ER, Eau Gallie River; MC, Melbourne Causeway; PC, Pineda Causeway; SI, Sebastian Inlet; TC, Turkey Creek. Grant is just west of Station 29. No. 4, 1982] OWEN AND NORRIS — GONYAULAX CYSTS 229 bucket which was then filled with water from the collection site for transport to the laboratory. The volume of the sediment sample was deter- mined after settling. By comparing this volume with the area of the mouth of the grab (21.6 cm x 23.5 cm) in the open position and assuming it sampled a uniform depth upon closing, the depth of the grab penetrated into the sedi- ment at each station was calculated. Samples in the jars were oven dried and subjected to the sieve analysis technique outlined in Lambe (1951). Sieve analysis data were used in an at- tempt to relate the distribution of G. monilata cysts to the depositional en- vironment of the area. Samples in the bucket were sieved to concentrate the cysts and remove all particulates that did not fall between 20 /on and 90 fim. The concentrated samples were stored in the dark at 7°C until examined; a minimum of three, 0.1 ml subsamples were examined at 100 x magnification for G. monilata cysts on a counting slide (Palmer and Maloney, 1954). Additional sub- samples (2 to 3 from Stations 15 and 16 and as many as 10 from each of the other 22 stations where no cysts were observed) were analyzed to verify our initial findings. Thus, over 1 ml of concentrated sample from most stations was examined. The results of this effort did not alter the findings of the ini- tial 3 subsamples as reported below. Our results are reported simply as cysts present or absent. Due to agitation of the sediments by wind, boating activ- ity, and bioturbation, we feel any cysts in at least the upper 1 cm would be capable of excysting and thus releasing a motile cell into the water column. Cysts as deep as 10-15 cm may be important as seed populations (Anderson, pers. comm.). A few cysts were isolated into enriched lagoonal water to check their viability and if possible their identification. Only 2 of 10 cysts excysted after 5 da, neither survived to a recognizable motile stage. However, because the cysts counted in this study had the same general appearance, we assumed they were all viable G. monilata cysts. Walker and Steidinger (1979) described G. monilata cysts, their Fig. 21 is representative of the cysts from Indian River sediments. Results — Each station was sampled once during the 10 wk sampling period; depth at the stations ranged 0.6-5. 3m. Water temperature ranged from 14.2°C on 10 February to 25.0°C during the second week of April. Salinity ranged from low (ca. 20 %o) winter values during February and early March to the high (ca. 34 % ) values of late March and April (Owen, 1979). The sediment types encountered in the Indian River ranged from poorly- sorted sediments with large percentages of particles in the pebble, granule, very coarse sand, coarse sand and medium sand sizes to well-sorted sediments made up almost exclusively of medium or fine sands (Table 1). There appeared to be no definite relationship between the cyst distribution and the type of sediments found. Gonyaulax monilata cysts were seen in the sediments from 13 of 37 sta- *.'

230 FLORIDA SCIENTIST [Vol. 45

Table 1. Indian River stations with G. monilata cysts and their sediment descriptions.

The depth of grab is the grab penetration into the sediments assuming it samples a uniform depth upon closing (revised from Owen, 1979).

Water Depth Depth of grab Station (m) (cm) Sediment Description

2 1.7 2.9 poorly sorted, coarse to fine sands 4 1.3 2.4 poorly sorted, coarse to medium sands 6 2.0 5.3 poorly sorted, medium to fine sands 7 2.0 2.4 poorly sorted, very coarse to medium sands 10 1.0 3.1 well sorted, medium sands 11 1.1 2.8 well sorted, medium sands 15 4.0 4.6 poorly sorted, very coarse to fine sands 16 2.0 2.2 poorly sorted, coarse to fine sands 18 0.8 1.6 well sorted, medium sands 22 0.6 5.2 well sorted, medium sands 24 2.0 3.3 well sorted, medium sands 30 1.1 1.7 well sorted, fine sands 33 1.6 2.5 well sorted, medium sands

tions (Fig. 1, Table 1), most regularly on the east side of the Indian River between the Pineda and Melbourne causeways. In this section, they were found at all the eastern sites except Station 14. South of the Melbourne

Causeway, cysts were found at only 7 of 23 stations; 4 on the western and 1 on the eastern side of the lagoon, 1 (Station 15) adjacent to the intra-coastal waterway, and 1 (Station 30) just east of Grant Island.

Discussion — Advection in the Indian River is little more than a sloshing back and forth in the individual basins with a restricted mixing between ad- jacent basins due to causeways. Also, areas within a given basin, formed by sand bars and spoil islands, are essentially isolated from the general circula- tion pattern (Trees, 1977). This limited advection, both within and between basins, in large part accounts for the spatial distribution of benthic G. monilata cysts. Blooms of this toxic dinoflagellate have been reported in the Indian River between the Pineda and Eau Gallie causeways, between the Eau Gallie and Melbourne causeways, and the basin just south of the Melbourne Causeway (B.C.H.D., unpublished data; Modert, 1977; Trees, 1977). Like those of the Galveston, Texas area (Wardel et al., 1975), these blooms usually occurred in July and August (sometimes September) when the water temperature rose above 29 °C and the salinity was over 32 %o. Of the 13 sites where G. monilata cysts were found, 9 are within the areas where blooms have developed in the past (Stations 2, 4, 6, 7, 10, 11, 15, 16, 18). Stations 2, 4, 6 and 7 are in the general area where fish kills were recorded by the B.C.H.D. during the summers of 1973, 1975, and 1978 (Ray Grizzle, pers. comm.). Though no specific identifications or cell counts were made during these fish

kills, dinoflagellates were present in high numbers in red water areas which had moderate dissolved oxygen concentrations (> 3 ppm). Streaks of "white" and green water with lower dissolved oxygen (sometimes < 1 ppm) NO. 4, 1982] OWEN AND NORRIS — GONYAULAX CYSTS 231

were also present. Unfortunately, the B.C.H.D. saw no evidence of whether the fish were actually killed in the red water or the low oxygen areas. The bloom reported by Trees (1977), which formed between the Eau Gallie and Melbourne causeways, provides some evidence supporting the assertion that the cysts remain in the area of the outbreak. The bloom was first seen at the end of June 1977 with concentrations between, 4,000 and

1 30,000 cells •l" . Concentrations of G. monilata in this area increased steadily during early July and on the last sample date (July 12) had reached 5 5 1 between 1 x 10 and 8.9 x 10 cells •l" . The bloom occurred in the northern and central areas of the basin that included Stations 9, 10, and 11, of this study. Benthic cysts were found at Stations 10 and 11, (Fig. 1, Table 1), none was seen at Station 9 probably because sediment deposition from the Eau Gallie River has buried them. Cysts were not found in the sediments col- lected at stations (i.e., 12, 13, and 14) in this basin which were not affected by the 1977 bloom. The restricted water movement that allowed the bloom to remain in the northern portion of the basin also accounts for the presence of the cysts in the sediments of the area. The hypnozygotes or cysts were able to settle out of the water column before they could be transported any great distance. Perhaps similar to that described for tide pools in Tokyo Bay (Hirano, 1967) or more recently for salt ponds in Cape Cod (Anderson and Wall, 1978; Anderson and Morel, 1979a, b), the method of bloom development in the Indian River can be understood in terms of the weak circulation and the distribution of cysts in the sediments. The first G. monilata red tide was recorded in 1951 in the Indian and (adjacent) Banana rivers (Howell, 1953).

After 30 yr, blooms of this dinoflagellate still occur when conditions become optimal. Walker and Steidinger (1979) proposed that cysts in Florida waters act as seed beds to initiate G. monilata blooms. The density of benthic cysts in the Indian and Banana rivers can apparently provide the 4,000 to 6,000 motile cells •l" 1 of water that are commonly seen in the initial stages of a bloom. After excystment, the dinoflagellates reproduce asexually and the bloom continues to grow for as long as conditions can support it. Because there is only very weak, limited water circulation and the wind has no major affect on the advection of the water and associated dissolved and/or sus- pended matter (Trees, 1977; Meyer, 1977), the dinoflagellates remain and become concentrated in limited areas. As is generally believed when condi- tions become unfavorable, cysts form and accumulate in the sediments serv- ing as seeds for future outbreaks. In addition, there appears to be no background level of motile cells in this lagoon; thus the cysts are apparently the only source for the seed popula- tion of G. monilata blooms. David (1978) did not find motile G. monilata cells in a 6-mo study covering the area south of the Pineda Causeway to just south of the Melbourne Causeway. This apparent lack of motile cells of bloom species in the plankton for periods of months was also reported by Hirano (1967). 232 FLORIDA SCIENTIST [Vol. 45

In the southern part of the study area where there are no records of past G. monilata blooms, G. monilata cysts were found at 4 sites, Stations 22, 24, 30 and 33 (Fig. 1). The latter 3 sites are in areas leased to commercial shell fishermen. Although these are in regions which appear to lack the widespread distribution of cysts that occur in the bloom areas to the north, localized blooms are possible in them. This is especially true at Stations 24 and 33 because of an apparent restricted water circulation. The areas around both sites are separated from the rest of the basin by a series of spoil islands and sand bars (K.C.O., pers. obser.). Such lack of water circulation might allow G. monilata to reach concentrations that are toxic to fish (Gates and Wilson, 1960; Ray and Aldrich, 1967; Sievers, 1969) and to a number of invertebrates including Crassostrea virginica, the American oyster (Sievers, 1969). Field observations have provided conflicting reports of the effects of G. monilata blooms on Crassostrea virginica. Wardle et al. (1975) reported no

6 1 effects on C. virginica during a 1971 bloom (1.2 x 10 cells •l" ) while there was "common" occurrence of "moribund" C. virginica during a 1972 bloom

6 1 (1.88 x 10 cells •l" ) along the beachfront at Galveston, Texas. Walker (un- publ. data) found no shellfish mortality during a G. monilata bloom (5.7 x 6 -1 10 cells ^l ) in August 1979 in Pensacola Bay, Florida. Also, no mortalities of marine organisms were observed associated with the bloom (maximum

7 1 cell count of 1.65 x 10 cells ^l" ) in Mississippi waters in August 1979 (Perry et al., 1979). Thus, there is not sufficient field evidence to implicate direct toxicity due to G. monilata blooms. G. monilata is not a causative agent of paralytic shellfish poisoning. The presence of cysts in such close association with the shellfish beds around Grant appears to be more than just coincidental. For several years, shellfishermen have been relaying large quantities of oysters from the Banana River and sections of the Indian River north of the Eau Gallie Causeway to their lease areas near Grant (Capt. Henry Morgan, Florida Marine Patrol, pers. comm.). Because red tides caused by G. monilata blooms are known to occur in areas from which oysters are transplanted, it is reasonable to assume that oysters rafted to the leases have had G. monilata cysts associated with them. Thus, they would be introduced into these areas in the same manner as has been suggested for the colonization of new regions in Cape Cod, Massachusetts by G. tamarensis (Anderson and Wall, 1978). We emphasize that G. monilata has not been associated with shellfish toxic- ity to humans. Acknowledgments — We thank Raymond Petrilla and Kurt VanGelder for help in collecting sediment samples; Linda Walker, Karen Steidinger, and Donald Anderson, for critical review and suggestions; and, Thelma Coughlin for typing the manuscript.

LITERATURE CITED

Anderson, D. M. 1980. Effects of temperature conditioning on development and germination 16:166-172. of Gonyaulax tamarensis (Dinophyceae) hypnozygotes. J. of Phycology. of two red tide blooms by the germina- , and F. M. M. Morel. 1979a. The seeding NO. 4, 1982] OWEN AND NORRIS — GONYAULAX CYSTS 233

tion of benthic Gonyaulax tamarensis hypnocysts. Estuarine and Coastal Mar. Sci. 8:279- 293. 1979b. Toxic dinoflagellate blooms in the Cape Cod region of Massachusetts. Pp. 145-150. In: Taylor, D. L., and Seliger, H. H. (eds.). Toxic dinoflagellate blooms. Elsevier, New York. and D. Wall. 1978. Potential importance of benthic cysts of Gonyaulax tama-

rensis and G. excavata in initiating toxic dinoflagellate blooms. J. of Phvcology. 14:224- 234. Dale, B. 1977. Cysts of the toxic red-tide dinoflagellate Gonyaulax excavata (Braarud) Balech from Oslofjorden, Norway. Sarsia. 63:29-34.

David, J. R. 1978. Net-phytoplankton ecology of the Indian River: the effect of light intensity and temperature upon standing crop during the period October 1977 to April 1978. M. S. thesis. Florida Inst, of Tech., Melbourne.

Gates, J., and W. B. Wilson. 1960. The toxicity of Gonyaulax monilata Howell to Mugil cephalus. Limnol. and Ocean. 5:171-174. Haddad, K. D., and K. L. Carder. 1979. Oceanic intrusion: One possible initiation mechanism of red tide blooms on the West Coast of Florida. Pp. 269-274. In: Taylor, D. L., and Seliger, H. H. (eds.). Toxic dinoflagellate blooms. Elsevier, New York. Hirano, R. 1967. Mechanism of development of red tide in estuarine waters. Information Bulletin on Planktology in Japan. Commemoration Number of Dr. Y. Matsue. 42:25-29.

Howell, J. F. 1953. Gonyaulax monilata, sp. nov., the causative dinoflagellate of a red tide on the east coast of Florida in August-September, 1951. Trans, of the Amer. Microscopi- cal Soc. 72:153-156. Lambe, T. W. 1951. Soil testing for engineers. John Wiley & Sons, Inc. New York. Meyer, R. G. 1977. A description of wind tides, astronomical tides and circulation in the In- dian River near Melbourne, Florida. M. S. thesis. Florida Inst, of Tech., Melbourne. Modert, C. 1977. The effects of ultraviolet radiation on the production rates of natural popu- lations of estuarine phytoplankton. M. S. thesis. Florida Inst, of Tech., Melbourne. Owen, K. C. 1979. Description and distribution of dinoflagellate cysts in the sediments of the Indian River, Brevard County, Florida. M.S. thesis. Florida Inst, of Tech., Melbourne. Palmer, C. M., and T. E. Maloney. 1954. A new counting slide for nanoplankton. Amer. Soc. Limnol. and Ocean., Spec. Publ. No. 21. Perry, H. M., K. C. Stuck, and H. D. Howse. 1979. First record of a bloom of Gonyaulax monilata in coastal waters of Mississippi. Gulf Res. Rep. 6:313-316. Prakash, A. 1967. Growth and toxicity of a marine dinoflagellate, Gonyaulax tamarensis.

J. Fisheries Res. Board of Canada. 24:1589-1606. Ray, S., and D. Aldrich. 1967. Ecological interactions of toxic dinoflagellates and molluscs in the Gulf of Mexico. Pp. 75-83. In: Russell, F. E. and Saunders, P. R. (eds.). Animal toxins. Pergamon Press, Elmsford, New York. Sievers, A. M. 1969. Comparative toxicity of Gonyaulax monilata and Gymnodium breve to

annelids, crustaceans, molluscs and a fish. J. of Protozoology. 16:401-404. Steidinger, K. 1973. Phytoplankton ecology: A conceptual review based on eastern Gulf of Mex- ico research. CRC Critical Reviews in Microbiology. 3:49-68. 1975a. Basic factors influencing red tides. Pp. 153-162. In: LoCicero, V. R. (ed.). Proceedings of the First International Conference on toxic dinoflagellate blooms. Mas- sachusetts Sci. and Tech. Found., Wakefield, Massachusetts. 1975b. Implications of dinoflagellate life cycles on initiation of Gymnodinium breve red tides. Envir. Letters. 9:129-139. Trees, C. 1977. Suspended solids in a basin of the lagoonal system of Florida. M. S. thesis. Flor- ida Inst, of Tech., Melbourne. Walker, L., and K. Steidinger. 1979. Sexual reproduction in the toxic dinoflagellate Gonyaulax

monilata. J. Phvcology. 15:312-315. Wall, D. 1975. Taxonomy and cysts of red tide dinoflagellates. Pp. 249-255. In: LoCicero, V. R. (ed.). Proceedings of the First International Conference on toxic dinoflagellate blooms. Massachusetts Sci. and Tech. Found. Wakefield, Massachusetts. Wardle, W., S. Ray, and A. Aldrich. 1975. Mortality of marine organisms associated with offshore summer blooms of the toxic dinoflagellate Gonyaulax monilata Howell at Gal- veston, Texas. Pp. 257-263. In: LoCicero, V. R., (ed.). Proceedings of the First Inter- national Conference on toxic dinoflagellate blooms. Massachusetts Sci. and Tech. Found. Wakefield, Massachusetts. Florida Sci. 45(4):227-233. 1982.