ARE RED TIDES CORRELATED TO SPRING-NEAP TIDAL MIXING?

:
USE OF A HISTORICAL RECORD TO TEST MECHANISMS
RESPONSIBLE FOR DINOFLAGELLATE

BLOOMS

W.M. Balch
Institute of Marine Resources, A-018
Scripps Institution of Oceanography
University of California, San Diego
La Jolla, CA

92093

INTRODUCTION

Vertical mixing
influential
(Sverdrup,
plankton

factors

is generally
affecting

1953; Gran

abundance

and

considered

phytoplankton

Braarud,

Changes

and species composition

at the sources of water

mixing

considerable

tidal

one

attention.

stratification
depth,

is

parameter,

a frictional
current

of

current

Research has therefore been

the

are important

turbulence

(C )

and

the

tidal

Most

Hunter

which

has

current

the

( M ) tidal
(M )

and

turbulence

formulated

cubed

appears

to

applications

lunar

(M )

as well.

For

semi-diurnal
the

third

sensitive to changes

currents but

fortnightly

the mean

is quite

of

(1974)

received

is a log function of water column

Since

in creating

dynamics.

and

velocities.

that

ocean
phyto

Simpson

have concerned semi-diurnal

evidence

the

both

most

S, which

power, the stratification parameter

tidal

in

in

the

column turbulence in nature and how

source

parameter

(U).

of

are thought to occur as a

these sources affect phytoplankton community

Tidal

one

abundance

1935).

function of turbulence (Margalef, 1 9 7 8 ) .

directed

to be

of

this

in

model

there

is growing

tidal

components

example,

Simpson

and Bowers (1981) have shown with models that thermal fronts in shelf
seas would be expected to move on a fortnightly cycle. Simpson
used

satellite

greatest

two

imagery

days

Webb and D'Elia

to

demonstrate

following

spring

(1980) showed

that

were vertically

homogeneous

Chesapeake

(see

sediments
(Postma,

Bay
has

1967;

suggested

that

also

also

been

Balch
such

et

tidal

spring

1977).

observed
al.,

to

1983).

mixing

frontal

in

nutrient

during
Haas,

that

tides

advancement

European

shelf

and oxygen
tides

in

Resuspension
occur

during

Holligan

events might

(1981)

and
affect

was

waters.

distributions
tributary
of

of

estuarine

spring
Harbour

tides
(1977)

phytoplankton

abundance.

Lecture N o t e s o n Coastal and Estuarine Studies, V o l . 17

Tidal Mixing and Plankton D y n a m i c s . Edited by J. B o w m a n . M. Yentsch and W. T. P e t e r s o n
Springer-Verlag Berlin Heidelberg 1986

Copyright American Geophysical Union. Transferred from Springer-Verlag in June 1992.

194

The

prediction

correct.
et

al.

(1975)

(their

tidal cycle on
and

of

primary

using

also

strong

observed
Lawrence

the surface

and M

show

effect

the

between

primary

Balch

and

saw

and community

that

neap

different

statistically
and

density

Moreover, he

to the

in

the

spring-neap

similar

observations

in

period

fluctuations

in

lunar

structure

waters

algae

of

phytoplankton

at

was

answered

the

spring

mandala

spring

nutrients

stratified

tide),

are

then,

depleted,

and

turbulence

of

with

water

a spring

Chesapeake

is "Do
phases

1978,

Bay.

bloom

1979)

term

was

arose

layer

(i.e., just

event

after

Haas e_t al.

heterolobatum

in a tributary

to

describe

dominance of one species of phytoplankton over all other species.

frequently, but

not

intensified
cycle.

exclusively,
as

Diatom

involves water

stratification
blooms

resumed

coincident

with

discoloration.)

during
the

the

there was a

be favored.

is used

will

in which

low but

Cochlodinium

"bloom"

stratify

dinoflagellates

tide destratification
(The

as waters

If the condition

might

red

detail

specific

(Margalef,

tides,

then dinoflagellates

observed

associated

neap

M^

seas and estuaries, diatoms will be favored

during

and

in great
during

high concentration of nutrients in the upper

bloom

All of these observations

predominate

Margalef's

in shallow

tides

column

the

spring-neap

bloom dynamics can be affected by

increase in relative abundance.

of

Winter

production

to be related

(1981)

proven

concentration

production

in the Gulf of Maine.

cycle"?

in well-mixed

(1980)

the

found a

in biomass

appeared

phytoplankton

lunar

predicts

of

chlorophyll

(1978) also

One question which has not been

water

has

tidal components,

during

an

stability,

Sinclair

that

while

phylogenetically
of

(1977)

detail, the data of

Haas e_t a_l. (1980) made

abundance

that

5)

fluctuations

Estuary

shelf break station

suggest

Harbour

the data of Winter et al_. (1975).

tidal variability.
Chesapeake

and

in great

column

correlation

stratification

the

Figure

the water

production.

significant

St.

Holligan

Although not discussed

the

spring

the
It
The

subsequent
tides

and

dinoflagellate blooms following the spring tides were observed in the

Gulf of Maine (Balch,

Questions
abundance
historical

of

1981).

concerning
algal

record

taxa
of

red

the
can

effect
be

water

of

lunar

answered,
blooms.

dinoflagellates) have been well documented

in

Red

periodicity
part,

tides

because

by

on

the

using

the

(usually

due

to

1) they are often

easily observed, 2) they are often associated with fish mortality and

195

human

affliction

followed

by

(Taylor

massive

Sanders, 1 9 5 7 ) .
cally

up

after.

to

and

algal

Seliger,

die-off

Dinoflagellate

the mid-19th

through

Margalef's
waters

the

and

during

tidal

dinoflagellate

neap

tides.

according to specific

quite

are

documented

well

often

(Brongersmasporadi

documented

there
In this

(1) dinoflagellate red tides occur

spring-neap

mandala)

3) they

in coverage and span 150 years.

paper I will test the hypotheses:

evenly

and

putrification

blooms have been

century

The data are global

1979)

and

cycle,

red

will

and

tides

discuss

(2)

occur

(relating
in

segregating

coastal regions and particular

to

stratified
the

species

data

respon

sible for the blooms.

METHODS

Data on red tides were collected from published accounts as well

as from a polling of dinoflagellate ecologists around the world.
date

of

first

sighting,

the location,
Nautical

the

species

responsible

for

(taken

Almanac, Washington, D.C.) were tabulated.

was calculated by setting the date of the new moon

day after a new moon as day 1.
days in length and
any given month
mid-day,

it varies

(Nautical

then the

than 29.

lunar

units

The lunar phase

The lunar cycle is approximately

several hours about

Almanac).

cycle

can

"days"

were

exact

the
29.5

this mean value

in

If the lunar cycle begins after

encompass

used

when

30

integer

days

and all data concerning

rather

specified

specified

comparing

occurred to the phase of the lunar cycle.

fore not

the

as day 29 and

(unlike the lunar cycles which were

of

water,

from

Due to the fact that bloom observations were only

to a given day
minute)

the red

and the date of the previous new moon

The

the

day

This calculation

to the
a

bloom

is there

the phase of the lunar

cycle

during which a bloom occurred have a confidence interval of + 1 day.

The
data

as

influence
29

tabulated

day

the

circular

occurred

eliminated

of

on

lunar

cycle

histograms.

day

30

from the data set

of

the

was
Two

out

lunar

(this did not

rejection of the null hypothesis).

examined
of

cycle.
affect

by plotting
the

Southern

by

comparing

California

continental

the

Bight,

total

the

shelf, New Guinea,

Gulf

Chile

data
of

set

to

Mexico,

were

the acceptance

or

against

Kolmogorov-Smirnov

test and statistical tables (according the Zar, 1 9 7 4 ) .

examined

blooms

data

Distributions were tested

even distributions using the Chi square test, the

was

226

These

the

the

Regional bias
data

the

(20-30 latitude),

W.
and

from

the

European
Gulf

of

196

Maine/New
was,

England

however,

Smirnov

waters.

The data base for the specific

too small to use the Chi square test. The

("KS")

test

remains

robust

at

small

sample

locations

Kolmogorov-

size,

however

(Zar, 1 9 7 4 ) , so the data were pooled into six 4 day intervals and one
5

day

test

interval

and

tested

against

an

even distribution with the KS

(the single 5 day interval was weighted appropriately to make it

comparable to the 4 day interval.)

The Rayleigh test

used to check for a significant mean direction

grams.

The median

(Zar, 1974) was

in the circular histo

angle was also calculated

(Zar, 1974) which, when

compared with the mean angle, allows detection of symmetrical

butions versus asymmetrical

A 62 year

record

distri

distributions.

of daily

sea surface temperature

anomalies

at

the Scripps pier was examined for lunar periodicity using time series
averaging
formed

(Bendat

and

Piersol,

1971).

All manipulations

were

per

with an HP9845B computer. The anomaly was calculated by aver

aging the daily temperatures over 62 years and subtracting each daily
temperature from the 62 year average for that day (e.g. mean temp for
Nov.

5 over 62 years - temp

1938).

Time

anomaly

for

December

series
each

for Nov.

averaging

day

of

the

5, 1938 = anomaly

involved

last

62

fitting

years

for

the

(January

Nov.

5,

temperature
1920

through

1981) to a day of the lunar cycle, where day 1 was the day

after

a new moon.

lunar

cycle

were

periodicity.

Then

It was

anomalies greater

all

averaged
also

or

of

and

the
the

possible

less than

anomalies
29

to

day

for

record

scan

the

62

each
was

of

the

examined

day

for

year

record

for

some designated temperature, and note

their timing during the lunar cycle.

RESULTS

The data set used

were

226

observations

mortality
received

or human

in this study
of

red

affliction

after this writing

in the calculations).

is given

tides

and

related

and are

to

in Appendix

62
red

observations
tides

(more

There

of

fish

data

were

included in the appendix but not

Figure 1 shows a world map with

sightings noted by their numerical

1.

designations

apparent that the sampling distribution

locations

(Appendix 1 ) .

is not uniform.

of

It is

5 9 . 4 % of the

observations were made from North American shores, 1 2 . 1 % from Central

and South America,

8.6%

from

Western

Europe

and

Iceland,

1.4%

from

the Black Sea, 2 . 4 % from Japan, 4.5% from India, 5.2% from Africa and
6.2%

from

the

Phillipines

and

New

Guinea.

Figure

shows

the

197

198

GLOBAL

RECORD
0 2 4 6 8 1012 14
1 i

NO.

Figure 2. 29

day

tide

circular

sightings

global

in

histogram
through

coverage.

showing

the

lunar

Phase

perimeter of the histogram.

of

OF

i i .

BLOOMS

the distribution
cycle.

the

moon

The
is

data
noted

of red
set
on

is
the

The scale appears in the lower

left.

circular

histogram

of

red

7.83

deviation

tide

=2.87,

for the complete

observations
n = 29 d a y s ) .

(Kolmogorov-Smirnov

per

data set.
day

There was

an

of the lunar cycle

The data were normally

average

(standard

distributed

test against a normal curve, P < 0 . 0 5 ) . Chi square

and Kolmogorov-Smirnov

tests showed

this distribution

of blooms over

199

the

lunar

cycle

distribution
there

was

not

at an

no

to

be

alpha

significantly

effort

significant

of

direction

different

0.05.

Using

to

circular

the

from

the

an

Rayleigh

even
test,

distribution

in

Fig. 2 (P<0.05).

These data were treated separately to examine any regionality in

bloom timing.

Figure 3 shows circular

6 geographic regions.
data into 4 day

increments

these distributions
bution
some

(KS

test,

evidence

3B)

showed

except

in

are significantly

P<0.05)

of

each location.

lunar

although

bias

in

red

Southern

California.

different

from

from

qualitative

tides

that

None

examination

was

bimodal

distribution

with

the W. European

5 0 % of

not

shows

the same at

the

cycle

blooms

(28% of

(Fig.

observed

the

cycle).

Shelf, 3 2 % of the observations were made

between

days 14 and 17 of the lunar cycle

lunar

of

an even distri

For example, the Gulf of Maine/New England data

from days 2-5 or days 18-21 of the

On

histograms for red tides

The small sample size necessitated pooling the

(10% of the c y c l e ) .

Forty

percent

of the Gulf of Mexico blooms were documented between days 27 and 1 or

between

days

6 and

of

the

lunar

cycle

(28% of

the

cycle).

In

Southern California, 1 7 % of the blooms occurred on days 28 and 29 (7%

of the lunar c y c l e ) .

Figure

observations
lunar

shows
were

cycle.

locations.

The
Red

the

data

plotted
tidal

tides

broken

against
ranges

clearly

down

tidal
varied

occur

in

by

region

range

but

rather

considerably
regions

of

the

than

bloom

day

of

amongst

these

variable

tidal

range (0.2 up to 4 m ) .

Proper
that

interpretation

of

Figure

a very high or low tidal range

is

complicated

is seen

by

the

less frequently

than an

average tidal range (due to the sinusoidal nature of the tidal

over a fortnightly

tidal

cycle).

Therefore,

fact

range

if red tides occur ran

domly with respect to tidal range, then more blooms would

occur

when

there was an average tidal range than during those few occasions when
the tidal range was very high or low and we might mistakenly

conclude

that there was a relationship between tidal range and red tide forma
tion.
"number

The
of

data

in

Figure

were

replotted

blooms" had been weighted

(not

by the frequency

of a given tidal range, but no relation between

tides was observed.

shown)

tidal

of

after

the

occurrence

range

and

red

200

Figure 3. Circular histograms showing the distribution of red tide

sightings through the lunar cycle at six general locations
around the globe: a) New Guinea (n = 1 5 ) ; b) Gulf of Maine
and New England waters (n = 1 4 ) ; c) S. California Bight (n
= 6 0 ) ; d ) Chile (20-30S latitude, n = 1 4 ) ; e) Gulf of
Mexico (n = 3 5 ) ; and f) the W. European Shelf (n = 2 2 ) .
All histograms but that for S. California Bight involved
small data sets which necessitated pooling the data into
six-4 day increments and one-5 day increment.
The number
of observations in the 5 day increment was weighted by 4/5
to make it comparable to the other 4 day increments.

Southern California Bight

As

already

Observations

illustrated

in

Figures

3C

and

4C,

little correlation to the lunar phase or tidal range

California Bight.
the

same

appear
results

region

to
of

anomalies

be
the

red

tides

in the

show

Southern

Nevertheless, the 62 year temperature record from

shows

evidence

correlated
time

from the

to

series

the

of

temperature

lunar

averaging

62 year record.

cycle.
of

perturbations
Figure

Scripps

Pier

which

shows

the

temperature

It is apparent that there was a

201

TIDAL RANGE ( m )

Figure 4. Histograms of number of bloom observations v s . tidal range

for the 6 regions given in Fig. 3.
"Tidal range" refers to
the maximum tidal range observed on the day of a reported
bloom.
Arrows on the abscissa designate the minimum and
maximum tidal range for each particular locality (given in
the U.S. Department of Commerce tidal t a b l e s ) .

25 2 7 2 9 I

II

13 15 17 19 21 2 3

DAY O F L U N A R C Y C L E
Figure 5. Time series averages of SIO pier temperature
anomalies
through the lunar cycle.
Data are from a 62 year daily
record of sea surface temperatures.
Lines on either side
of the central average line represent 1 standard error.
The lunar phase is designated below the abscissa.

202

cyclic

nature

cycles.
erature

to

the

average

temperature

anomaly

through

The general trend of the anomaly was that

anomalies

led

the

lunar

cycle

the

lunar

the maximum

temp

by 0 to 2 days.

The maximum

peak to trough difference in the anomalies was 0.196C. Highest

eratures were observed at day
were

seen at days 5-6

27 and day 12 and coldest

and 20.

There

the two peaks at days 27 and 29.

different
of

is no

obvious

temperature

successive

anomalies

is

explanation

However, they are not

from the anomalies of the surrounding

analyzed

the

of

for

significantly

days.

using

"troughs," it shows a frequency

temp

temperatures

If the
period

2 cycles

cycle

between

per

29

days

(or 0.069 cycles/day).

In any sampling
frequencies
sampled

must

every

semidiurnal

of

be

24

a time

series,

considered.

hours.

The

aliasing

The

data

SIO

were

tidal cycle of temperature

by

pier

checked

other

important

temperatures
to

see

were

whether

(1.935 cpd; Pond and Pickard,

1978) sampled every 24 hours could appear as a fortnightly cycle with

a frequency

of 0.069 cpd.

The equation of Bendat and Piersol

was used to calculate the alias frequencies

a

frequency

of

0.069

cpd.

The

closest

(1971)

and none were found

harmonic

to 0.069 cpd

with
is at

0.0645 cpd (15.5 days per c y c l e ) .

Figure
lies

in

the

relative

shows

evidence

Southern

that

large negative temperature

California

Bight

to the lunar tidal cycle.

have

occurred

The general

trend

anoma

non-randomly
in the histo

grams is cyclic and the maximum cold

(>-4C) anomalies lags the lunar

cycle

(Fig.

by

approximately

-4C were most

and

least

1 to

between

28 of the lunar cycle.

days

7-10

Figure 6A

and

day

significantly

days

7-14

and

-3C were most

6B).

Anomalies

exceeding

25-27 of the lunar cycle.

commonly

seen on days 20

The -3C anomalies were least frequently

24 of

is significantly

the Chi square test

not

days

frequent on days 1-5 and days 16-20 of the lunar cycle

frequent

The anomalies exceeding

on

the

lunar

different

cycle.

The

seen

distribution

in

from an even distribution using

(P<0.05) while the distribution

different

and

in Figure 6B

from an even distribution

(P<0.05).

is
The

two histograms are similar in shape, however.

Several

major

blooms

on

the

Southern

California

continental

shelf can be defined as cell concentrations sufficient to yield 50 ug

chlorophyll

a J!,''" (the minimum

ble red water discoloration

chlorophyll

is about 5 ug

a concentration for visi

; Holmes et al. , 1 9 6 7 ) .

203

, 9
I
I

u
o
r<')
I
II

I
I
I
I
I

Z
<t
:x::
I

riO

I
I

I
I
I
I

u
0

C)
I
I
I

I
I
I
I

1\

<t
::E
0
z
<t

:x:: 5
I

(/)

(/)

o
u...
o
d
z

0
u...
0
0
d
z
DAY OF THE LUNAR CYCLE

A.

B.

Figure 6. Histograms

showing

the

number

of

observations

of

desig

nated temperature
the phase of the

anomaly in the S.
California Bight vs.
lunar cycle.
Panels A and B refer to
o
0
negative temperature anomalies of >_3 C and >_4 C, respec
Phase of the moon is also shown below the X axis.
tively.
Data are from the 62
year daily record of sea
surface
temperature at the SIO

total

of

(Table 1;
place

imply

at

these

major

blooms

in Appendix 1).
after

the SIO

an

pier

have

Each

abrupt

(Fig.

of

fall

7).

been
these,

in

sea

recorded
save

since

one,

surface

has

1920
taken

temperature

These sharp temperature declines

nutrient-rich surface waters since several nutrient concentra

tions vs.

0
temperature relations have shown waters colder than 14-15 C

contain

1983)
kelp

of

immediately

measured

to

refs.

pier.

measurable

except
beds

in

nitrate

while

warmer

special locations with

(Jackson,

1977).

(If

we

waters

do

not

(Jackson,

impeded circulation as within

were

sure

the

same

water

parcel

was observed over time we could say the blooms followed mixing events
when

the

waters

were

illustrated in Fig.
the

S.

California

once

again

6A and B,
Bight

becoming

stratified.

As

already

major negative temperature anomalies in

tend

to

lag

the

new

and

full

moon

by

1-4

days.)

The
answer

entire
the

data

question,

set

was

"Are

analyzed

there

with

single

respect

species

to

species

to

which bloom at a

204

CD

au

CD CD
Ci SH X )
CD CD S

T3

ft<H

r>oo<Hor>rHoc>cn
N

n o) n a oi

CM

o, cd a
< u

CM CM

T3

rH
>> >>
X i 4->
ft-ri
O CD

CD

13
CD

SH SH

3
3
CO CO
oi oi
CD CD

u a
O (D-<
rH T ) l
XI
CJ S

3 t
:

a a
o o o m o c o o o o
m O O O O I O C D r H O C M
A
rH
ITJ
H H M<

<D S
P -rH
at X
cd

8
ri a
x
O -P
SH oS
ft

a
<

CO
(D
ri
CJ
CD

RJRJ

oi

oi

SH

CO

SH

X!
CD

ho

&

ci

o,

ois

HNS

ri

ci

N
rH
|D O
CD
ftl
X
ri

blftioioiftia oi

m
O

SH CD
Oi r H
>> C O
a) 3 >>
Q rH O

73

CD

>

SH

CD
CD CO
P X I
o! O
Q
P
CO
SH
ri

COrHr>^t<CS>CQCOCOCM
CM
CM
r H CM

ntoiO'#tooHH/ir)
coco^inmcocDcoco
0 ) 0 0 1 0 0 0 0 0 ) 0
r H r H r H r H r H r H H r H r H
>>>>-PrH
bO CD SH > > CD
o i o i f t o i 3 B a o i c i

SS

0>S<! 3 < J S 3

CO

c~ o
r H CO CO
rH

>-s

oi m

i-D

co o

rH O rH rH OJ
CO
I

205

Figure 7. SIO pier temperature record for two years in which major
red tides were observed (in which chlorophyll levels were
>50 pg H ) .
The solid black line represents the recorded
temperature and the dotted line represents the 62 year
daily average temperature.
Arrows denote the dates that
the major red tides were first observed.
l

particular

phase

Prorocentrum
times

that

of

micans,
red

the
and

lunar

Gymnodinium

tides were

tides of Gymnodinium
of the lunar cycle

cycle?"

breve

Only

breve

observed.

Gonyaulax

showed

Forty-five

any

bias

percent

(n = 24) were observed

polyedra,
in

the

of the red

from day 25 to 29

(17% of the total c y c l e ) , 2 4 % of the Prorocentrum

red tides (n = 17) were observed on day 28 of the lunar cycle

(3% of

the total cycle) and 4 0 % of the Gonyaulax polyedra red tides (n = 25)
were observed

from day 25 through day 1 of

the total c y c l e ) .

the

lunar

cycle

(21% of

There were not enough data of the other 25 species

to reliably examine for a lunar bias.

The annual cycle of dinoflagellate red tides is shown in Fig. 8;

northern

hemisphere

hemisphere blooms

blooms

are usually observed during

with

5 6 % of

southern

the

are

shown

in

in the lower panel.

summer months

observations

hemisphere,

the

red

between
tides

the

upper

panel,

in the northern

June

showed

and
a

hemisphere

September.
bimodal

In

the

distribution

through the annual cycle with 6 8 % of them occurring between

and May.

southern

It is apparent that red tides

December

206

NORTHERN

A.

HEMISPHERE

35
30|
25'
20'
15
10
5
J F M A M J

J A S O ND
SOUTHERN

B.

HEMISPHERE

n
12
II
I0|

9
8
7-

6
O

3
2

J F M A M J J A S O N D
MONTH
Figure 8. Histograms
showing
the
annual
cycle
of
dinoflagellate
blooming in a) the northern hemisphere, and b ) the southern
hemisphere.
Letters under the abscissa refer to months of
the year.

DISCUSSION

Two types of mechanisms can be invoked to explain

red

tides;

involves
(e.g.

are not mutually

a growth

response

dinoflagellate

vertical
sort

they

of

migration,
physical

to

excystment
etc.) while

concentration

Combinations of these mechanisms

exclusive.

some
in

exogenous
response

a physical
of

the

cells

dinoflagellate

A biological
or
to

(e.g.

factor

temperature

mechanism

are particularly

mechanism

endogenous

involves

cue,
some

Langmuir

cells).

appealing,

such as

207

physical turbulence transporting nutrients into the mixed layer and a

subsequent growth response of the algae, or stratification
vertical

migration

across

the

thermocline

concentration

(Kamykowski, 1 9 7 9 ) .

concentrating

mechanisms

nutrient
column

can

be

before

found

the

must

in

the

bloom

However,

operate
cells

in

than

(Ketchum

and

leading
it

is known

some

blooms

was

1948;

diel

horizontal

the

physical

because

present

Keen,

with

to

in

more

the

Holmes

water

e_t

al.,

1967).

Most
cycle

workers

and

who

have

phytoplankton

observed

blooming

relation

have

between

noted

that

the

lunar

spring

tidal

currents increase the thickness of the bottom mixed layer (or surface
mixed

layer;

seasonal

see

Hayward

thermocline

and

e_t

al. ,

inject

this

volume),

nutrients

into

layer (Pingree e_t al_. , 1976, for example).

during neap
tive

tides could give

advantage

stratification

if

a)

the

Subsequent

down

surface

the
mixed

stratification

the red tide dinoflagellates a competi

diatoms

of the water

break

fail

column,

to

remain

suspended

following

or b) if the dinoflagellates can

migrate into a deepening nutrient-rich layer at night and return to a

nutrient-poor
1975;

surface

Harrison,

showed,

1976;

however,

European

that

blooms

Nevertheless,

of

for

layer

during

Haas

et

such

the

al.,

vertical

Prorocentrum
migrating

day

(Eppley

1980).
migration

minimum

and

Paasche
was

and

dinoflagellate

et

not

Harrison,
al.

(1984)

required

Gyrodinium
species,

for

aureolum.

the

original

conceptual model formulated to describe the expected results for this

study on a global basis

is shown

tidal amplitude at new and

lies (on average) associated

3) positive
tides
more

and

anomalies

4) most

stratified

of

neap

in Figure 9A.

full moon,
with

average)

the

dinof lagellate
The

temperature

anoma

the increased spring tidal mixing,

(on

tides.

It predicts 1) peak

2) negative

during

the more

blooms

latter

is

stratified

associated

not

supported

neap

with
by

the
the

global view of Figure 2 so this conceptual model must be revised.

Pooling

the

data

into

a global

data

set

may

be

presumptuous,

particularly if one is invoking a tidal mechanism of bloom

Tides vary globally

as a function

of bottom

topography,

formation.

basin

reso

nance and general geographic location, not only with the phase of the
moon

(Pond

and

Pickard,

1978).

that the peak tidal range

full moon but
of

the

Garret

is not

and

Munk

(1971)

lags it by one or two days. They coined

tide" which

pointed

always synchronous with

refers to the lag between full or

out

the new or

the term

"age

new moon and

Figure 9. A.
Idealized conceptual model depicting the appearance of
dinoflagellate blooms through the lunar cycle (based on the
Margalef m a n d a l a ) .
The circular histogram is a representa
tion of what would be expected if dinoflagellates bloomed
during stable neap tidal periods.
This model incorporates
1 ) peak tidal amplitude at new and full moon, 2) negative
temperature anomalies associated with the increased spring
tidal mixing and 3) positive temperature anomalies asso
ciated with more stratified neap tides.
Lunar phase is
shown at the perimeter of the circular histogram.
B.
Data from the Southern California Bight to compare to
idealized conceptual model.
Data from Figures 3 and 5 were
used to make this figure.
Tidal amplitude is shifted in
this figure according to Garret and Munk (1971).

peak tidal

amplitude. Thus, the rationale

gellate blooms
the

variance

then

this

from 6 specific

due

would

to

areas

geographical

have

argued

factors

for

for examining the dinofla-

(Figs. 3,4) was to see

could

physical

be

reduced.

component

of

whether
If so,
bloom

formation.

Figure 9B depicts the observed

and

tidal

Bight,
model

amplitudes

which can

from

be directly

in Figure 9A.

It

red tides, temperature

a specific

area,

the

anomalies

Southern

California

compared with the original

conceptual

is apparent that the days of the lunar cycle

209

with

the

occur

significantly

1-2

days

anomalies.
tidal

after

However,

cycle.

Perhaps

were

other

more

red
more
the

are

lunar

tidal

blooms

(i.e. the

When

water

is the

cycle,

and

since

28

closer

and

29)

temperature
the

lunar

association

of

the spring tides and

This

is opposite

is nevertheless

consistent,

in

anomaly

on

the temperature

column

The mechanism

however,

(days

throughout

anomalies with

in Figure 9A but

observed.

complicated,

tides

time-averaged

anomalies with the neaps.

model

positive

tides

red

the

surprising

part, with M a r g a l e f s mandala.

average,

of

of

temperature

the average negative

idealized

number

maximum

there

the average positive

the

high

the

is more

of

1) blooms

2) plotting

are

the

is,

stratified),

bloom

formation
observed

data

to

more

must

be

throughout

relative

to

tidal

range rather than lunar phase fails to lower the variance of the data
set

(the blooms are observed when the

tidal

range

is

low

and

high)

(see Fig. 4 ) .

Kamykowski

(1981a) argues

from horizontal
through

concentration

internal

wave

that

dinoflagellate

red

fields.

Depending

on

the

current velocities, basin dimensions and migrational

his simulated
to

the

populations

shore,

2-10

km

and

0.42

h with

a 24

form concentrated

apart.

relies on superposition

of

tides

of dinoflagellates migrating

bands

Essentially,

internal

h diurnal

wave

the

differences

of

cells

of

tide
21.5

Conceivably,

lunar internal tides are important

parallel
formation

h, 12.5

pattern of dinof lagellate

through a baroclinic water column.

in

characteristics,

red

periods

result

diurnally

the fortnightly

in this context as well.

migration
and

Kamykow

ski 's model (1979; 1981a) is very appealing since several of the most
frequently observed

red tide organisms

in California are strong ver

tical migrators (Gonyaulax polyedra, Prorocentrum micans, Gymnodinium

splendens) (Kamykowski, 1 9 8 1 b ) .
breve,

Florida,

very

frequent

cause

It is not
of

red

is a vertical migrator.

It

known whether

tides

off

the

Gymnodinium

west

coast

of

is curious that G. polyedra, P.

micans, and G. breve show a strong lunar bias in bloom formation

(see

Results section).

Given
that

the

preceding

"dinoflagellate

spring-neap tidal cycle"

2).
by

results,

it

tides

are

red

cannot

The hypothesis cannot

geographic

evidence

for

region,
a

lunar

be

albeit

the

clear

observed

negated

be rejected

influence.

is

on

that

the

evenly
a global

hypothesis

through
basis

the
(Fig.

even if the data are analyzed

data
These

in

Figure

results

3
for

suggest

some

bloom-forming

210

dinof lagellates are

Sinclair
that

(1978),

observed

in contrast

Haas e_t al.

changes

to the data of Winter et al. (1975),

(1980),

Balch

(1981) and

other

in abundance of phytoplankton, not

large-celled dinoflagellates, related

workers

necessarily

to the lunar tidal cycle.

The

reasons for the discrepancy are not obvious but may be due to a suite
of factors which increase the variance of the data set.

Perhaps
from

the

"date

the

fact

of

that

first

previous

day,
it

successional

amount

the

data

sighting,"

observations,

Moreover,

largest

assumes
yet
is

there

stage

are

based

although

that

known

of variance

the

bloom

might
that

the

not

red

in the data set

on

human

best

way

was

have

observation.

not

been

tides

between

observable

on

observer

of

red

the

present.

represent

later

in the growth of dinoflagellate populations.

initiation of a bloom may occur weeks previous to the

vation

The

to compare

an

usually

results

tide.

Alternatively,

red

initial

tides

could

into a region which might create an "observational

lag."

be

The

obser

advected

One solution to these problems might be to intensively sample at

one sight.
of

When

bloom

cycle.

this

still

Nonetheless,

frequently
surface

is done

observation

observed

in

(e.g. La Jolla, CA, U S A ) , the

does

this

one

during that

temperature

anomaly

not

correlate

region

part

is, on

1)

well

red

to

pattern

the

tides

tidal

are

most

of the lunar cycle when the sea

average,

positive,

and

2) major

red tides (with chlorophyll a levels >50 ug/Jl) are often observed
days

after

warming.

large

negative

temperature

Perhaps the combination

stratification

provides

the

each year without bloom

Variance
discerned.
migrators
Thus

the

due

to

they

water

1-2
is

but

not

always

sufficient,

formation.

the

can

mechanism

the

Such events take place several times

wide

variety

of

species

Not all species of dinoflagellates

yet

when

of a sharp mixing event and slight

necessary,

conditions for bloom formation.

spike

form

surface

responsible

slicks

for

involved

can

be

appear to be vertical

(Paasche

dinoflagellate

e_t al_. , 1 9 8 4 ) .
aggregations

proposed by Kamykowski

(1979) is probably only one of several explan

ations for red tides.

Other purely physical concentration mechanisms

were

reviewed

by

Ryther

(1955).

The

fact

that

a non-motile

diatom

can form surface slicks (see Appendix, reference 6) is tribute to the

fact

that

such

physical mechanisms are important

Species specificity

in bloom timing

is further

in bloom

implied

dynamics.

when the data

211

are

segregated

(1940)

taxonomically

observed

species

(see

Results

specific

section).

seasonality

in

Indeed,

the

Allen

appearance

of

dinoflagellates in the Southern California Bight.

One
that

conclusion

phytoplankton

Phytoplankton
factors,
rates,

formation

species

grazing

tidal

obvious

"red tides" cannot

bloom

namely,

and

lunar

is particularly

is

potential

cycles,

as

by

migrational
well

semidiurnal

this

work.

It

is

be treated as a single entity.

affected

specific

from

as

host

of

biological

capabilities,

physical

factors

tides,

ephemeral

internal

growth
such

as

mixing

events, seasonality, etc.

Historical
prediction,
merely

during

quite

that

useful

toxic

the

red

each

8).

many

years

of

the months

Recognition
ago

by

of

of

resulted

the

that

the

the

of

prediction.
is

in

red

tide

For

example,

the

northern

1 3 - 1 4 % of the red tides

July,

August
in

"mussel

Department

a larger data base, predictive

context

concern

seasonality

in

California

of

June,

the

in

tide

locality

allows one to predict

Fig.

imposed

are

particularly

knowing

hemisphere

Given

data

of

occur

and September

toxic

quarantine"
Health

(see

dinoflagellates

in

each

summer

S.

California.

abilities might be

sufficiently

enhanced to allow species specific bloom forecasting

on shorter

time

scales than seasons.

Factors such as the previous temperature

record

are clearly important

for predicting

parti

cular

species

phases

of

show

the

a greater

lunar

cycle

large blooms.

likelihood

(e.g.

one

of

Moreover,

blooming

quarter

of

during

the

certain

Prorocentrum

blooms occurred on day 28 of the lunar c y c l e ) .

One

goal

techniques

of

and

relationship

this

symposium

observational

between

is to

strategies

physical

biomass and production.

series

structure,

outline

needed

the

to

types

validate

circulation

and

of
the

mixing

to

Future

studies on red tide formation

should

be more temporally comprehensive

in order to catch the initial

stages

of

bloom

formation.

This

is neither easy nor cheap since red

tides

are very sporadic in time and space (which is why I used the histori
cal

data

base).

Efforts

should

bias since this undoubtedly

Remote

sensing

adequate

water

difficult

to

provides

sampling

is

sample,

standing them remains

the
to

be made to remove the

observational

increases the variance of the

partial
still

most
study

solution
required.

reasonable
the

specific

to

this

Since
approach
red

tide

data

problem

red

tides

towards

set.
but
are

under

organisms

and

212

their distributions and natural history.

examined

in

broader
bloom

relation

insights

dynamics.

to

into
The

the

the

average

physical

paper

of

Their abundance can then be

physical
forcing

Yentsch

et

environment

functions

al.

(this

to

allow

important
volume)

to

is

an

example of this approach.

SUMMARY

An historical, global record of dinoflagellate red tides is used

to

examine

whether

dinoflagellate

lunar (M^) or fortnightly

with respect

blooms

are

correlated

Red

tides

are

not

as well

as

tidal

well

correlated,

on

global

that

at one

These

results

location,

are made

the Southern

or

The

regional

more

surprising

California Bight,

temperature record shows evidence for a lunar cycle of

and destratification.
associated
in

range.

location.

to lunar phase or tidal range, in contrast to the findings of

some other workers.

fact

the

to the phase of the moon

data are analyzed with respect to taxa and geographic

basis,

with

( M ) tidal components. Blooms are tabulated

this

Large negative

with particular

region

temperature

(>50

anomaly

ug

temperature

phases of the lunar

chl

suggesting

- 1

the

usually

the

stratification

anomalies are also

cycle.

follow

importance

by

a 62 year

of

Major
large

blooms

negative

upwelled

water

to

bloom formation. These historical data suggest that several different

mechanisms
should

are

focus

probably

responsible

on the natural

for

history

Future

work

of specific organisms and

red

tides.

their

typical physical environments.

ACKNOWLEDGEMENTS

Many
work.
red

people

Thanks
tides

temperature
Scripps
pier

to

have

go

to

me.

tation.

and

D.

and

much

provided

Eppley

along

with Dr. C M .

SIO

of

valuable

region

and

the

completion

elsewhere
provided

assistance.

Marine

Enright,

provided

in

kindly

computer

Institute

in the S. California Bight

at

Stewart

Drs. J.R.

Lang

Dr. R.W.

researchers

and

temperatures.

Jackson

instrumental

Betsy

data

Aquarium

been

on

took

this

reporting
SIO

Members

Eppley,

discussion

for
the

Resources

R.W.

of

pier

of

the

the
daily

L.W. Haas,
data

G.A.

interpre

data and ideas on the major blooms

and kindly

reviewed

Yentsch and two reviewers.

the manuscript

M.A. Ogle typed

the

213

manuscript.

W.M.B.

is

supported

by

NSF

grants

OCE81-20773

and

OCE80-08308.

REFERENCES

Allen, W.E.
1940.
Twenty years of statistical studies of marine
plankton dinoflagellates of southern California.
Amer. Midi.
Nat. 26: 603-635.
Balch, W.M.
1981.
An apparent lunar tidal cycle of phytoplankton
blooming and community succession in the Gulf of Maine. J. exp.
mar. Biol. Ecol. 55: 65-77.
Balch, W.M., P.C. Reed and S.C. Surrey-Gent.
1983.
Spatial and
temporal variability of dinoflagellate cyst abundance in a tidal
estuary.
Can. J. Fish. Aquat. Sci. (Suppl.) 40: 244-261.
Bendat, J.S. and A.G. Piersol.
1971.
Random data:
analysis and
measurement procedures. Wiley-Interscience, New York. 407 p.
Brongersma-Sanders, M. 1957.
Soc. Am. 6 7 : 941-1010.

Mass mortality in the sea.

Mem. Geol.

Eppley, R.W. and W.G. Harrison.

1975.
Physiological ecology by
Gonyaulax polyedra, a red water dinoflagellate off
Southern
California/ p! TT-22.
I_n
V.R. LoCicero ( e d . ) .
The First
International Conference on Toxic Dinoflagellate Blooms. Mass.
Sci. Techn. Found., Wakefield, MA, USA.
Garrett, C.J.R. and W. Munk. 1971. The age of the tide and the
of the ocean. Deep-Sea Res. 1 8 : 493-503.

"Q"

Gran, H.H. and T. Braarud.

1935. A quantitative study of the phyto
plankton in the Bay of Fundy and the Gulf of Maine including
observations on hydrography, chemistry, and turbidity.
J. Biol.
Bd. Can. 1: 219-467.
Haas, L.W.
1977. The effect of the spring-neap tidal cycle on the
vertical salinity structure of the James, York and Rappahannock
Rivers, Virginia, U.S.A.
Est. Coast. Mar. Sci. 5: 485-496.
Haas,

L.W.,
S.J. Hastings and K.L. Webb.
1980.
Phytoplankton
response to a stratification-mixing cycle in the York River
Estuary during late summer.
Tn
B.J. Neilson and L.E. Crown
(eds.).
Estuaries
and
Nutrients.
The Humana
Press
Inc.,
Clifton, NJ.

Harrison, W.G. 1976. Nitrate metabolism of the red tide dinoflagellate Gonyaulax polyedra.
J. exp. mar. Biol. Ecol. 21: 199-209.
Hayward, D., L.W. Haas, J.D. Boon, III, K.L. Webb and K.K. Friedland.
A regression model of neap-spring tidally associated stratifi
cation variation in the York River estuary, Maryland.
I_n
M.
Bowman, C M . Yentsch and W . T . Peterson (eds.), Tidal Mixing and
Plankton Dynamics.
Springer-Verlag.
This volume.

214

Holligan, P.M. and D.S. Harbour.

1977.
Vertical distribution and
succession of phytoplankton in the western English Channel.
J.
mar. biol. Assn. U.K. 5: 1075-1093.
Holmes, R.W., P.M. Williams and R.W. Eppley.
1967.
Red water
Jolla Bay, 1964-1966.
Limnol. Oceanogr. 1 2 : 503-512.
Jackson, G.A.
1977. Nutrients and production
cyst is pyrifera off southern California.
979-995.

in La

of giant kelp, MacroLimnol. Oceanogr. 22:

Jackson, G.A.
1983. The physical and chemical environment of a kelp
community, pp. 11-37.
In
W. Bascom (ed.).
The Effects of
Waste Disposal on Kelp Community.
Southern California Coastal
Water Research Project, Long Beach, CA.
328 p.
Kamykowski, D.
1979.
The growth response of a model Gymnodinium
splendens in stationary and wavy water columns.
Mar. Biol. 50:
289-303.
Kamykowski, D.
1981a.
The simulation of a southern California red
tide using characteristics of a simultaneously-measured internal
wave field.
Ecological Modeling 1 2 : 253-265.
Kamykowski, D.
1981b.
Laboratory experiments on the diurnal verti
cal migration of marine dinoflagellates through
temperature
gradients. Mar. Biol. 6 2 : 57-64.
Ketchum, B. and D.J. Keen.
1948. Unusual phosphorous concentrations
in the Florida "Red Tide" seawater.
J. Mar. Res. 7 ( 1 ) : 17-21.
Margalef, R.
1978. Life-forms of phytoplankton as survival alterna
tives in an unstable environment.
Oceanol. Acta 1: 493-509.
Margalef, R.
1979.
Functional morphology of organisms involved in
red tides as adapted to decaying turbulence, pp. 89-94.
Ta
D.L.
Taylor and H.H. Seliger
(eds.).
Toxic
Dinoflagellate
Blooms.
Elsevier/North Holland.
Nautical Almanac, 1900-1983.
United States Nautical
U.S. Naval Observatory, Washington, D.C.

Almanac

Office,

Paasche, E., I. Bryceson and K. Tangen.

1984.
Interspecific varia
tion in dark nitrogen uptake by dinoflagellates.
J. Phycol. 20:
394-401.
Pingree, R.D., P.M. Holligan, G.T. Mardell and R.N. Head.
1976. The
influence of physical stability on spring, summer and autumn
phytoplankton blooms in the Celtic Sea.
J. mar. biol. Assn.
U.K. 56: 845-873.
Pond, S. and G.L. Pickard.
1978.
Introductory dynamic
Pergamon Press, New York. 241 p.

oceanography.

Postma,
H.
1967.
Sediment
transport
and sedimentation
estuarine environment, pp. 158-179.
Ill
G.H. Lauff
Amer. Assoc. Adv. Sci., Publication 83.

in the
(ed.).

215

Ryther, J.H.
1955.
Ecology of autotrophic marine dinoflagellates
with reference to red water conditions, pp. 387-414.
I_n
F.H.
Johnson ( e d . ) .
The Luminescence of Biological Systems.
Amer.
Assoc. Adv. Sci., Washington, D.C.
Simpson, J.H.
existence
531-546.

1981.
The shelf-sea fronts:
and behavior.
Phil. Trans.

implications of
R. Soc. Lond.

Simpson, J.H. and D. Bowers.

1981.
Models of stratification
frontal movement in shelf seas. Deep-Sea Res. 28: 727-738.
Simpson, J.H. and J.R. Hunter.
Nature 250: 404-406.

1974.

Fronts

in

the

Irish

their
A302:

and

Sea.

Sinclair, M.
1978.
Summer phytoplankton variability in the lower
St. Lawrence Estuary.
J. Fish. Res. Bd. Can. 35: 1171-1185.
Sverdrup, H.U.
1953.
On conditions for the vernal blooming of
phytoplankton.
J. Cons. Perm.
Internatl. Explor. Mer. 18:
287-295.
Taylor, D.L.
Blooms.
U.S.

and H.H. Seliger. (eds.)

1979.
Elsevier/North Holland, New York.

Toxic Dinoflagellate
505 p.

Dept. Commerce, NOAA National Ocean Survey.

Tide tables high
and low water prediction.
U.S. Govt. Printing O f c , Washington,
D.C.

Webb, K.L. and C F . D'Elia.

1980.
Nutrient and oxygen redistribu
tion during a spring neap tidal cycle in a temperate estuary.
Science 207: 983-985.
Winter, D.F., K. Banse and G . C Anderson.
1975.
The dynamics of
phytoplankton blooms in Puget Sound, a fjord in the Northwestern
United States. Mar. Biol. 29: 139-176.
Yentsch, C M . , P.M. Holligan, W.M. Balch and A. Tvirbutas. Tidal
stirring
vs.
stratification:
Phytoplankton
dynamics
with
special reference to toxic dinof lagellates.
I_n M. Bowman, C M .
Yentsch and W.T. Peterson (eds.).
Tidal Mixing and Plankton
Dynamics. Springer-Verlag.
This volume.
Zar,

J.H.
1974.
620 p.

Biostatistical

analysis.

Prentice-Hall

Inc., N.J.

216

REF
t

RED TIDE
OR
POISONING

REFERENCE

LOCATION

YEAS

DATE
FIRST
SITED

DATE OF
PREVIOUS
NEW
MOON

DAT
LUNAR
CYCLE

SPECIES
OF
DESCRIPTION

La J o l l a , CA
U.S.A.

1917

4-VI

20-V

15

P. mlcans

I-VI

3-V

30

A l l e n , W . E . , Anr.
Midland N J C .
26:603-635

Allen, W.E., S.I.O.

B u l l . Tech S e r i e s
No. 1: 3*7-356

192*

W.F. A l l e n , Naut.
G a z e t t e , 26 Aug 1939

1939

W.E. A l l e n , Trans o f
Aider. H i C r o p , Soc.
62: 262-264

1942

1 1-IX

10-IX

G. p o l y e d r a

W.E. A l l e n , Science,
88:55-56

Copalis Beach, WA
U.S.A.

1938

14-V

30-IV

14

Aulacodlscus
klttoni

5
6

Seaside, WA,
N. Columbia R i v e r

1938

15-V

30-IV

15

Aaterionella
Kariana

La J o l l a , CA
U.S.A.

1938

Later
May

29-V

1?

C-. p o l y e d r a

W.E. A l l e n , T r a n s ,
o f Aaer. Microp.
Soe. 65= 149-153

19*5

-13-IX

6-rx

C. p o l y e d r a

W.E. A l l e n , Science,
78:12-13 (1933)

1933

17-V

2 4-IV

23

C. t r i p o s
P. mlcans

1935

30-VII

30-VII

29

Yellow f l a g e l

1898

8-IX

10

W.E. A l l e n , Science,
82:325-326 (1935)

11

Mead, Science,
8:707-709 ( 1898)

Providence, R . I .
U.S.A.

12

India

1949

27-X

13

T o r r e y , Am. Nat.
36: 187-192 ( 1902)

San Pedro Harbor

U.S.A.

1902

7-VII

5-VII

Gonyaulax s p .

IK

C u l l e n e_ a l - , J - E x p .
Mar. B i o l . E c o l .
63:67-80 (1982)

La J o l l a , CA
U.S.A.

1980

21-VII

12-VII

Yellow t i d e

15

W. Balch - p . O . T

1981

30-1

27-X

16

W. Balch and L .
Vakaaian. P . O .

1981

10-XI

27-X

14

17

W. Balch - P.O.

Del Mar, CA
U.S.A.

1981

26-XI

26-XI

29

18

D. Goodaan - P . O .

La J o l l a , CA
U.S.A.

1982

It-I

17-1

16

(Thattywar,

P e r l d i n l u B sp

G. polyedra?

19

W. Balch - P . O .

1982

19-1

17-1

20

W. Balch, H, Coven P.O.

80 k S. o f
T i j u a n a , Mexico

1982

25-1

17-1

21

J,. N e l s o n , W. Balch

La J o l l a , CA
U.S.A.

1982

30-1

17-1

13

22

Finucane $ i a l . , F l a .
Board o f C o n s e r v a t i o n
Rep.

Taapa Bay, PL
U.S.A.

1963

11-IV

25-III

17

C. brave

23

B. Sweeney, 1st I n t .
Conf. on T o x i c D i n o .
Blooas, Hasp. S c l . T e c h .
F o u n d a t i o n , 1975

La J o l l a , CA
U.S.A.

1952

11-VII

22-VI

19

P. Mlcans

2tt

25

1951

9-III

5-III

195*

26-IV

3-IV

23

1955

21-IV

24-111

28

1958

IT-VI

17-VI

29

26

27

28

195B

4-VII

17-VI

17

29

I m p e r i a l Beach,
CA.y.S.A.

1958

21-VII

30-YII

30

La J o l l a , CA
U.S.A.

1958

15-VIII

15-VIII

29

31

1961

13-IV

16-III

28

P. alcana

32

1961

28-IV

15-IV

13

0. p o l y e d r a

33

O y s t e r Bay,
Janaloa

1966

13-1

22-xii

22

P. bahanense

34

B o s t r a i n Bay,
New Guinea

1969

13-X

11-X

C. polygramaa

Ensanada, Mexico

0. p o l y e d r a

217

RE?
1

35

REFERENCE

RED T I D E
OH
POISONING

DATE OF
PREVIOUS
(JEW
MOON

DAY
LUNAR
CYCLE

U-X

30-IX

22-VII

22-VII

13

LOCATION

YEAR

DATE
FIRST
SITED

S a n t a M o n i c a CA
U.S.A.

1970

1971

30-VII

R I N O O N , CA
U.S.A.

1971

4-VIII

38

1971

7-VIII

22-VII

16

39

Oolata
U.S.A.

1971

16-VIII

22-VII

25

37

40

41

42

44

*5

46

47

B a y , CA

18-VIII

22-VII

27

8-IX

7-IX

0.

polyedra

1972

15-IX

7-IX

1973

22-X

26-IX

26

L o n g B A C O N , CA
U.S.A.

1974

2-VIII

19-VII

13

Ventura,
U.S.A.

197U

10-VIII

19-VII

21

Epplay 4 H a r r i s o n , l a t
I n t . C o n f . on T o x i c D l n o .
Blooaa, Mass. S c i . Tech.
F o u n d . 1975

1971
1972

SPECIES
OR
DESCRIPTION

CA

R i n o o n , CA
U.S.A.

1974

19-VIII

17-VIII

S e a l B e a c h , CA
U.S.A.

1967

3-IX

6-VIII

28

CEratiurn s p .

46

Sorrento Slough,
CA, U . S . A .

1968

15-11

29-1

17

49

La J o l l a ,
U.S.A.

1968

27-111

28-11

28

C.
P.

50

1968

17-V

27-IV

20

Ceratium

CA

polyedra
alcana
sp.

P. a l c a n a
51

52

53

54

55

56

57

58

59

60

62

63

64

65

66

1969

26-11

16-11

10

G.

polyedra

1969

27-V

16-V

1 1

P.

micans

NEWPORT B e a c h , CA
U.S.A.

1970

3-IX

31-VIII

La J o l l a ,
U.S.A

1970

22-X

30-IX

22

1970

9-XII

28-XI

11

1971

22-VIII

20-VIII

Mesodiniuto
C. f u r c a

1972

24-IV

13-IV

1 1

Dino.

C.

CA

La J o l l a ,
U.S.A.

Hartwell, lat I n t . Conf.

on T o x i c D l n o B l o o a a .
Haas. S c i . Tachn Pound.
1975

k Varlnnar

67

69

70

71

72

polyedra

1974

8-VIII

19-VII

20

1974

16-IX

16-IX

29

E s s e x , MA
U.S.A

1974

20-V

22-IV

28

1972

4-IX

9-VIII

26

"

1973

11-V

2-V

1973

16-V

2-V

14

1973

1 1-IX

28-VIII

14

1973

26-IX

26-IX

Ipswich
U.S.A.

C h e s a p e a k e Bay
U.S.A.

68

0.

El Segundo, L O S
A n g e l e s , CA, U . S . A .

Offshore
Bay, HA,
Zubkoff
Ibid

CA

polyedra

G.

tanarensis

1973

31-X

26-X

1974

21-11

23-1

29

1974

4-III

22-11

10

1974

27-111

23-HI

1974

11-IV

23-111

19

1974

7-V

22-IV

15

73

1974

24-V

21-V

74

1974

10-VI

21-V

20

75

1974

2-VII

20-VI

12

76

197l

12-VIII

19-VII

24

77

1971

16-IX

16-IX

29

78

Tantsch 1
Salvagio,

1972

14-IX

7-IX

I p s w i c h B a y , MA
Ibid

mixture

?
G.

tanerensis

218

REF

RED TIDE

BOOTHBAY H A R B O R ,
MA, U . S . A .
SELLGER,

YEAR

DATE
FIRST
SITED

DATE O F
PREVIOUS
NEW
MOO (I

1971

8-IX

17-VIII

22

U-VI

15

18-VI

2II-V

15

25-VIII

20-VIII

IBID

1971

ST. PETERSBURG,
PL, U . S . A .

JENSEN,

IBID

LONG L A . SOUND,
NY, U . S . A .

1972

12-VIII

9-VIII

197 3

22-X

26-IX

1973

3-IV

1972

7-IX

CAPE E L I Z A B E T H ,
HE, U.S.A.

30-V

21-V

HASPTON, N . H .
U.S.A.

21-VIII

17-VIII

1-IV

9-III

1-X

13-IX

1976

28-VI

27-VI

1977

F R A GA 4 S A N C H E Z ,
IBID
RIO DE V I G O ,
SPAIN

4-VIII

27-VII

30-IX

13-IXZ

1952

STI
THAT
OUTBREAKS
COINCIDED
WITH
PULL

14

MOON

14

1977
1977
IBID

ROBERTS,

BLASCO,

1976

IBID

GYMNODINIUM S P .

1976

1953

ZOTTER,

GONYAULAX
PERLDLNIUM
CERATLUN

1977

1954

1976

30-IV

SPECICS
OR
DESCRIPTION

3-IV

4-X

MACHADO, I N
"TOXIC
D L N O . BLOOMS" P R O C . O F
2ND I N T . CONF O F T O R I O
D I N O BLOOMS, E L S E V I E R N . H O L L A N D , MY 1 9 7 9

DAY
LUNAR
CYCLE

18-IV

12

16-VII

15

18-XII

21-XI

27

22-IX

25-VIII

28

23-IX

1977

13-IX

TANARENSLS

G.

TAAERENSIS

IN

31-VII

1976

C.

COCHLODINLUM

G . BREVE

IBID
10-IV

7-IV

5-VII

22-VI

20-VII

23-VI

P.

3-VIII

22-VII

C. FUSUS

1972

5-VII

1 1-VI

NOCTLLUCA M I L I A R I A

1972

21-VI

11-VI

1973

2-VII

30-VI

1973

18-IX

26-IX

1971

HURST, I B I D

HOLLIGAN,

IBID

P . MLCANA
REDFIELDII

7-VIII

19-VII

C . FUAUS

20-VI

20-VI

H. M I L I A R I A

8-IX

5-IX

1976

30-IV

29-IV

1976

30-VI

27-VI

MOOSE C O V E , ME

14-VII

27-VI

U A H A N T , W.
E N G . CHANNEL

31-VII

27-VII

SW M A I N E , U . S . A .

219

SEF

RED TIDE

DATE
FtPST

DATE OF
PREVIOUS

CAY
LUNAR

SPECIES
Cfi

SITED

NEW

CYCLE

CESCRIFTIC'N

MOON

Cepe Romano, FL
U.S.A.

23-XI
York R i v e r
Chesapeake,

U.S.A.

M i l l Creek,
Chesapeake,

U.S.A.

1976

1976
Clark,

Ibid

Muilr,

Ibid

Korey G a i n e s ,

White,

Ibid

Ibid

Balch, J . Exp.
Biol Ecol.

A. K r l J c o s .

Baker *

D.

Her.

P.O.

Dustan,

Redalje,

P.O.

G. K l e p p e l ,

P.O.

1980

13-X

12-X

16-VIII

27-VII

20

15-IX

13-IX

30-VIII

10-VIII

20

G.

nelson!1

Ceratluffl s p .

1982

25-VI

Adjacent to Elands
Bay, S . A f r i c a

1966

12-VI

11-VII

28-VI

18-XII

12-XII

16-XII

12-XII

C.

grlndleyii

1967

B a l l a n t i n e t S m i t h , Bi
Phycol. J . 8:233-238

138

139

15-11

Pt A r e n a To
P t . Reyes

136

137

14-VIII

25-VI

G r i n d l e y and N e l ,
S. Afr. D I v l . Fish.
B u l l . 6:36-55 M970)
135

27-VII

15-VIII

198.?

1980

12
1

2-VIII

1978

San D i e g o to S a n
F r a n a c i s o , U.S.A.

11-XI

27-VII

28-VII

Hiekel 4 Drebes, Helgo.

Wise. Meeresunteree
22:401-416

N. W a l e s ,
Conwy, U . K .

5-X

Llandudno,
U.K. ( ? )

5-X

N. S e a
Helgoland
Bight
30-VIII
Esbjerg

Pingree.

(N.

Sea)

22-IX

26-VII

9-VII

31-VIII

6-VIII

1968

20-IX

23-VIII

1967

10-VII

7-VII

7-VH

25-VI

Nature

258:672-677

(1975)

Ilzuka 4 I r l e , Bull.
Plank. S o c . Japan
1 6 : 9 9 - 1 1 5 C1969)

Onura B a y ,

UJeno, B u l l . Plank.
S o c . 16:89-98 ( 19*9)

1968

I l z u k a , B u l l . Plank.
Soc. japan 19:22 (1972)

Tangen,

23-VIII

14-X

Sarsla

63:128-133 (197?)
Brongersaa-Sandera,
G e o l . S o c . of America
67:941-1010 (1957)

21-U

23-VIII

1967

1-VIII

7-VII

1968(7)

7-VIII

25-VII

Stavenger,
Horwsy ( ? )

1976

15-XI

23-X

Brit.

1933

28-IV

Oaura Bay,
Japan

Columbia

Fort Bragg,

U.S.A.

Gynnodlniua s p .

C.

aureoliu

Meso,

1896

End Feb

1917

4-VI

1907

E a r l y Aug.

1938

5-IX

T938

12-IX

25-VriI

1946

19-VI

30-V

rub

220

YEAR

Angel de la Gardia
Is., Gulf of Cal.

1937

Chile

1695

Manila, Phllllpines

1767

Oulf of Maine,
U.S.A.

1862

DATE
FIRST
SITED

DATE OF
PREVIOUS
NEW
HO ON

20-111

12-111

DAY
SPECIES
LUNAR OR
CYCLE Dt'C:f!IPTr.

N. aclntlllana

7-II
22-IX
1-VIII

18-VI
1-VI
Iceland
Portugal
Zaton, Blaok Sea
Cape Blanoo,
Morocco, Afr.

S.W. Africa

1845

1913
1951
1944
1946
1950
1851
1880
1920
1925
1950
1945

Florida, U.S.A.

12-VII

Meaodln. rubrun

21-VII

3-VII

Yellow-red water

17-VIII

2-VIII

17-11

25-1

18-VII

1946
1946
1882
1946
1947

1947
1952
1952
1953
1953

3-VI

"Protococcua^

25-111
25-XII

3-III
9-XII

5-XII
21-XII
10-XII
23-XII

25-XII
24-1

9-XII
15-XII
9-XII
14-1

26-VIII

26-VIII

16-IV

9-IV

20-VII
24-X

20-XI

18-VI

20-VI
18-VII
25-X

18-VII

Kathiawer, N.W.
coaat India

Malabar,
W, India

Oulf of Manor,
India

1947
1936
1950
1849
1861
1916
1916
1916
1922
1944
1946
1946
1942
1931
19*0

29

18-1
18-X

24

11-XI
3-1

31-XII-52 3

22-1

15-1

10-XI
Yeaen, Red Sea

27

23-VIII

16-VIII

27-IX

15-IX

2U-X

11-1

27-X

"Last wk 28-VIII

August"
25-IX

28-VIII

9-X

26-X

22-IX

20-IX

3-XI

17-X

20-IX

26-VIII

31-X

24-X

17-V

15-V

5-XI

11-X

24-VII

5-VII

221

REF

DATE
FIRST
SITED

FED TIDE

e>stt. J. COD*. Int.

Exp. Mer 39:1-6

S.W. Africa

B. Balah * 0.
Redalje, P.O.

La Jolla, CA
U.S.A.

208
209

Rounsefell * Nelson,
U.S. Dept. or Interior,
Flan end Wildlife Svo.
Bureau Coa>. Fish Ho.

1880

21-XII

1982

4-IV

DATE OF
PREVIOUS
NEW
KOOtf

DAY
SPECIES
LUNA!! OR
CYCLE DESCRIPTION

G. polyedra

1982

22-IV

23-IV

1935

27-VI

1-VI

1953

22-1

15-1

1959

29-IX

3-IX

1959

22-X

2-X
25-HI

23-111

27

G. polyedra

535, 1966
210

Chew, Ibid. 1955

Dry Tortugas,
Caribbean

2IT

Funloaae, Ibid, I960

St. Peteraberg,
FL, U.S.A.

212

213

1963

3-IV

21k

Glennan, U.S. Fish

COBB. Bull. 1887

1885

28-X

215

Qunter, Sol 105:256

1947

19-1

216

Gunter, Eool.
Monogr. 18:309

19*6

20-11

9-XI

11

1947

2-IV

22-111

11

193*

Wid July

11-Vii

1954

24-XI

26-X

29

1966

23-111

26-11

25

196 6

27-HI

27-111

29

1880

17-1

1878

20-XI

1955

30-Hl

1953

18-IX

217
218

Hart, J. Nature
Nature 134:439

219

Hela. Bull. Mar.

Sci. Gulf
Carlb. 5:269

Dragovich,
Caribbean?

Huttoe,
Quart. Fla. Acad.
Sci. 23; 163

223

Jefferson, J.P.,
Ibid 1:363

224

Lackey, J.B.,
Quart. J. Fla.
Acad. Soi. 19:71

227
228

Dry Tortugas,
Caribbean

21-III
8-IX

18-iu

5-rn

13

1953

3-II

9-VTII

25

1954

18-VI

1-VI

17

Midnight Pass,
FL, U.S.A.

1953

18-XII

Sattlbal Light,
FL, U.S.A.

1954

Big Sarasota Bay,

FL. U.S.A.

12

229
230

232

Nuaann, Arob.
Flsoaerelwisa
8:204-209

1951

27-VII

4-VII

23

Poeveroy j i a^,,
Limnol. Oceanogr.
li54-60 (1956)

1955

21-IV

24-HI

28

Slobodkln, J.
Mar. Res. 12:148

1952

2-VI

23-V

10

197

20-VI

18-VI

21

1916

3-X

26-IX

1952

25-X

18-X

233

Fort Hyers, FL
U.S.A.

234
235
236

238
239

Saamer 1 Clark,
Cal. rieh and Oatve
32:100 (1946;

Santa Monloa,
CA, U.S.A.

1946

19-VI

30-V

20

Walker, Proo,
0.3. Rat. Mua.
6:105-109 (1884)

Birt rey, FL,

U.S.A.

1880

20-XI

30-V

20

Obua At Al-, Bull.

Mar. Sal. Cexib. 7955

Sarubel is.,
FL, U.S.A.

1954

30-IV

3-IV

27

Sarasota Pt.,
FL, U.S.A.

1954

12-VIII

29-VII

14

GynnodlniuH sp.

222

HEF

DATE

RED TIDE

DATE OF

DAY

T. PECI

FIRST

PREVIOUS

LUN,1F!

OF

IRI'tD

mew

cyci e

LE.-c;.tr:

Hoot:

Sanlbel I s . ,
FL, U.S.A.
B. Balch & S ,
Horrigan, P.O.
McLean, Pepeu New
Guinea A g r l . J o u r .
24:131-138

1951

23-VII

30-VI

1953

1B-IX

8-IX

1982

15-V

22-V

20-IV

T5-IV

T a l a s e a , Papua,
New Guinea

1961

Manus l a .
New Guinea

1962

23-V

U-V

1963

29-VI

21-VI

Port Moresby,
New Guinea

1970

16-XI

30-X

Lae, New Guinea

1971

7-IV

26-111

Marshall Lagoon,
New Guinea

1971

18-IV

26-111

1971

31-V

1972

13-IIR

15-H

1972

27-IV

13-IV

1972

12-V

13-IV

1973

15-IT

3-11

1973

20-11

3-11

Kapa Kapa t o Hood

P t . New Guinea

1973

9-III

5-IH

T a l a s e a , New

1973

R u l i g e r P t . , Milne
Bay, W. New B r i t a i n

197 3

Milne Bay,
New Guinea

1973

26-VI

1-VI

P. bahamense

P. b a h a n e n s e

1956

S. P l a c t e r , P . O .

18-IV

11-IV

1968

9-III

26-11

1972

22-X

1973

23-1

4-1

Aaphldiniun s p .

1975

10-11I

11-11

MeaodiniuM

Valparaiso, Chile

1975

9-III

11-11

Chaflaral,

Chile

1975

12-XII

3-XII

Me J i l i o n ,

Chile

1976

6-1

1-1

Antofagasta,
Chile

1976

25-11

Mejillon,

1976

22-IV

30-111

1976

7-IX

30-111

1976

15-V

29-IV

1976

28-X

23-1

1976

30-XI

1977

10-1

Valparaiso, Chile

Chile

Arloa, Chile

M e a o d l n i u B rubrum

G. c a t e n e l l a

rubru*

C. t r i p o s
P. micans

C. Turca

Gyanodiniun s p .

21-XII1976

10-1

Glenodlnlujo a p .

Ayaen, Chile

1978

16-111

9-III

Mesodinius rubrum

Valparaiso, Chile

1979

T-V

26-IV

P. nleans

1980

27-XI

7-XI

Iquique, Chile

1980

23-111

7-XII

Valparaiso, Chile

1981

8-IV

4-IV

Antofagasta,
Chile

223

HEF

REFERENCE

BED TIDE
OR
POISONING

Q u a y l e , D.B.
F i s h . Res. Bd. Can.
168:1-67

280

281

it

282

283

It

LOCATION

YEAR

B r i t i s h Columbia
Canada

British
Canada

Columbia,

1793

DATE
FIRST
SITED

DATE OF
PREVIOUS
NEW
MOON

DAY
LUNAR
CYCLE

SPECIES

on
DESCRIPTION

15-VI

1942

2-V

15-IV

1943

20-VII

2-VII

18

1948

1-IX

5-VIII

27

17

284

1949

13-IX

24-VIII

20

285

tt

1957

23-X

8-X

15

286

it

287

288

289

290

1963

12-VII

21-VI

21

1963

20-VI

23-V

28

1965

1-VI

30-V

1958

4-II

19-1

16

1943

12-III

6-III

291

292

P
P

293

ti
M

1963

18-XI

16-XI

1967

19-VI

8-VI

11

7
7

B. B a l c h &
F . T . H a x o , P.O.

La J o l l a , CA
U.S.A.

1983

9-IV

12-IV

25

T i e r r a d e l Fuego
S. America

1972

22-X

7-X

15

1972

11-XI

6-XI

22

294

Guzman & C a m p o d o n l c a ,
I n t e r c i e n c i a 3:144-151
(1972)

295

Guzman si. a l . , A p a r t ado

Analis del I n s t l t u t o
de l a P a t a g o n i a
6:173-183
n

ti

296

297

298

it

299

it

300

301

B. Sweeney &
A . D o d s o n , P.O.

Robinson & Brown,

Can. J . F i s h . A q u a t .
S c i . 40:2135-2143

G.

catenella

II

1972

28-XI

6-XI

1960

30-VI

24-VI

G.

ti

1964

10-V

12-IV

28

Mixed

ti

1965

1-VI

30-V

G.

1974

14-X

16-IX

28

1976

5-X

23-IX

12

1976

16-XI

23-X

24

1976

21-XII

21-XII

29

L a J o l l a , CA
U.S.A.

E a q u i m a l t Lagoon
J u a n de Fuoa
Strait

polyedra

species

polyedra
?

G.

sanguineus

302

303

304

1977

27-IX

13-IX

14

Gymnodinium s p .

305

"

1978

14-IX

2-IX

12

G. s a n g u i n e u m

306

tt

tt

1978

2-X

2-X

29

307

II

II

1979

15-X

21-IX

24

308

II

1980

24-VI

12-VI

12

309

II

1980

6-VIII

12-VII

25

310

II

311

II

T " P > 0 . " means " P e r s o n a l O b s e r v a t i o n "

Reads,

"...it

being

spring tide"

1980

20-VIII

10-VIII

10

1980

22-X

9-X

13

it
Gymnodinium s p .

t
M

G.

sanguineum