Results of the Territorial Monitoring Program

of American Samoa for 2007, Benthic Portion.

By Douglas Fenner, Ph.D.

Coral Reef Monitoring Ecologists

Department of Marine & Wildlife Resources (DMWR)
American Samoa

2008

Report to DMWR, the Coral Reef Advisory Group (CRAG), and NOAA

Supported by a NOAA Coral Reef Monitoring Grant, part of the Coral Reef Initiative

1
Contents
Acknowledgements 3
Abstract 3
Methods 5
Results 7
Benthic 7
Reef Slopes 7
Bioerosion and indicator species 19
Individual sites 20
Coral diversity in transects 32
Lifeforms 32
Generic and species diversity 35
Reef flats 43
Coral lifeforms 48
Coral genera 52
Coral species 53
Non-point pollution 55
Coral recruitment 56
Harbor 59
Bleaching 65
Invertebrates 67
Water clarity 69
Species 72
Recommendations 73
References 73

2
Acknowledgements
Thank you to boat captains, Lemuelu Kitiona and Mika Letuane. Thanks also to Ekueta
Schuster for filling tanks and other support. Special thanks go to the Bill Kiene of the
Fagatele Bay National Marine Sanctuary, for the use of the Sanctuary boat when the
DMWR boat was not available.

Abstract

One new site was added in the northwest, at Masacre Bay.

The mean benthic cover for all 12 sites was unchanged from 2005 to 2007. All
differences between years of all benthic categories were less than 3%, and due to random
chance. Thus, there was no loss of corals, but also no evidence of increase in corals.
This does not provide support for the view that the reefs are actively recovering from
previous disturbances, rather, they appear to be in a stable equilibrium. As in previous
years, crustose calcareous algae (CCA) had much higher average cover on the south side
than the north side, and higher turf cover on the north than the south. However, the new
site at Masacre Bay was much more like that on the south, with high CCA cover and
lower turf cover, illustrating that the rule is not absolute. The live coral index declined
slightly from 2005 to 2007, however it remains much higher than the average for the
South Pacific, and the decline may have been from more careful recording or from
random effects. All dead corals are old and covered with algae, usually coralline algae.
Coral cover on Tutuila was lower than reported at four remote unpopulated atolls in the
Line and Phoenix Islands, but more than at two populated atolls in the Line Islands. The
amount of coral plus CCA cover was much higher than on the two populated atolls in the
Line Islands, and nearly that of the unpopulated atoll in the Line Is. When coral cover +
CCA is plotted as one variable, and turf + fleshy macroalgae plotted as a second variable,
Fagasa and Vatia stand out as being quite different from the other sites in having low
values of the first, and high values of the second. This suggests that these two sites may
be more heavily impacted than other sites. Both appear to have significant sediment
impact. Trends for the three years were examined for each site. Fagamalo had a modest
increase in coral cover and a decrease in CCA. Fagasa had a modest increase in coral
cover, a decrease in turf, and an increase in fleshy macroalgae. Tafeu had an increase in
the corallimorph Rhodactis sp. (a large polyp related to corals that does not have a
skeleton) and a decrease in turf. Vatia had an increase in fleshy macroalgae and a
decrease in exposed sand. In 2007 the macroalga Dictyota appeared suddenly in large
quantities, growing over Halimeda, sand and live coral. This species had not been
recorded previously at any transect at any site in any year. It is one of the types of algae
that have grown over reefs in the Caribbean. Changes at Aoa, Aunuu, and Amaua were
small and not systematic. Fagaalu had a decrease in turf from 2005 to 2006, but no
change since then. Nuuuli had an increase in branching coralline algae (BCA) and a
decrease in CCA recorded. Likely the decrease in CCA was due to it being covered by
BCA, and it is still there under the BCA. Fagatele had a decrease in coral cover from the
first to the second year that may have been due to a change in the placement of the
transect tapes. In 2007 a small amount of the encrusting ascidian Diplosoma simile
appeared in the transects. This species is capable of growing over corals and killing
them. There was a report that it became quite common in one area of the bay, but then

3
receded in the following year. There was little change at Leone. Changes in the mean
cover of coral lifeforms in the transects at the 11 sites were small, with a slight reduction
in the amount of encrusting corals and increase in the branching (including columnar)
corals. There was a slight decrease in Montipora (which is usually encrusting) and slight
increase in Porites over the three years. There was little change in coral species other
than a small decrease in encrusting Montipora. The number of coral species in transects
did not change over the three years, nor did the diversity measure, H’. Monitoring of
reef flats was begun for the first time this year. Outer reef flats were dominated by turf,
and turf was the most common cover item on the inner reef flat. Coral cover was higher
on the outer reef flat than the inner, but turf was higher on the inner flat than the outer.
There was slightly more CCA on the outer flat than inner. When comparing the reef flat
with the slope, the slope had higher coral cover and much higher CCA cover than the flat,
but much less turf. A cohort of juvenile table corals (Acropora hyacinthus) were
observed. They were most common near the reef crest on both the reef flat and reef
slope. They were most common at Fagasa, where transects revealed 0.8 colonies per m2.
They were abundant enough that if they all survive to maturity they will produce near
100% coral cover in some areas. In most places they are much less common than Fagasa.
Sizes were measured and the distributions are essentially unimodal, suggesting a single
recruitment event. They appear to be about 3 years old. Transects were also taken at
several sites in the harbor on the outer reef flat or on rock walls. Coral cover was highest
near the mouth of the harbor where it reached high maximum levels (over 60%), but
decreased to zero near the head of the harbor. A small oyster is abundant near the
waterline on rock walls in the harbor, so estimates of the oyster cover were made for
several sites in the harbor. Oyster cover was high near the head of the harbor and low at
the mouth of the harbor. Oysters are filter feeders feeding on plankton and may be a
good bioindicator of plankton concentrations. Visibility was then measured in the harbor
with a Secci disc. Clarity was low near the head of the harbor, and high near the mouth.
Plankton concentrations are high at the head of the harbor, probably because of nutrient
inputs and very limited flushing, and low near the mouth of the harbor where flushing is
much better. Conditions near the head of the harbor apparently are still not good enough
to support coral. Coral bleaching monitoring continued, and mass coral bleaching of
staghorns in backreef pools was recorded for the fourth year in a row. At Alofau, these
staghorns spend almost the entire year bleached, with only a short period of recovery. So
far, most corals recover each year. The bleaching record at the airport pool follows the
sea surface temperature (SST) very closely. There continue to be relatively few visible
invertebrates.
In general, spatial patterns of benthic cover, both between sites (along the shore) and
in distance from shore (inner reef flat, outer flat, slope) showed strong variation and clear
patterns, while temporal patterns (changes over time) at the same location had little or no
variation at some sites, and strong changes at only a few sites, with no overall average
changes. Thus there are strong spatial variations, but little or no temporal variation so
far.
Results were illustrated with 89 figures. The report is 77 pages long.

4
Methods
The 11 core sites are shown in the map below. All are on Tutuila and nearby Aunu’u.

The benthic methods were the same as in 2006, with a few minor changes. In the
core monitoring, two 50-m tapes were laid on a depth contour between 8 and 10 m deep.
A space between them of about 15 m was kept. Benthic categories were recorded under
each 0.5 m point on the tape. Benthic categories included live coral, dead coral, dead
coral with algae, crustose calcareous algae, branching coralline algae, fleshy macroalgae,
turf algae, rock, sand, rubble, soft coral, and sponge. “Branching coralline algae”
included a soft feathery species that was the most common in that category. That species
is Cheilosporum spectabile. Corals were identified to lifeform, genus, and species when
possible, and if the macroalgae was Halimeda or Dictyota, that was recorded. Soft corals
were recorded to genus when possible. Lifeforms included encrusting, massive, foliose,
branching, columnar, submassive, mushroom, Millepora, Acropora branching, Acropora
table, Acropora digitate, and Acropora encrusting. Horizontal visibility was recorded
using the tape. Two transect tapes were done on the first dive, and an additional two
tapes were done on the second dive. Invertebrates were recorded on a return pass. Sites
were re-located using the GPS and markers as indicated in the 2005 report. One day was
required for each site. The same 11 core sites on Tutuila as recorded in 2005 were
repeated, which are shown in the map above.

5
 Changes include the rugosity measurements which were omitted, because a third team
member was not available and when included it lengthened dive times to the point where
running out of air was a distinct possibility, thus reducing the margin of safety. Further,
it appears that the measurement depends primarily on exactly where the chain falls, and
that changes in rugosity caused by coral growth will take quite a few years before they
would be detectable. A hurricane could make changes in rugosity quickly by removing
corals, and if significant hurricane damage occurs, the rugosity measurements can be
repeated. Until changes in coral cover or other rugosity changes are apparent, repeating
the measurement of rugosity is not worth the increased risk of running out of air. In
addition, the biodiversity measures were not repeated this year, due to the completion of
monitoring being greatly delayed by inclement weather and mechanical problems with
boats. It will be resumed next year.
In addition to the 11 core sites, one site was added at Masacre Bay on the northwest.
The two sites on the south (Amaua E and Funamafuti) that were added last year were not
repeated due to the lack of time. It is hoped that Amaua E can be added on a regular
basis in the future.

Dates of collection of data are shown in Table 1.

Table 1. Dates of collection of benthic transect data for each site, reef slope.
Transects
Fagamalo 3/6/08
Fagasa 1/24/08
Tafeu 2/28/08
Vatia 2/27/08
Aoa 2/25/08
Aunu’u 6/4/07
Amaua 5/18/07
Faga’alu 6/8/07
Nu’uuli 5/14/07
Fagatele 6/25/07
Leone 11/28/07
Masacre Bay 2/26/08

Table 2. Dates of collection of reef flat transect data for each site.
Fagamalo 3/29/07
Fagasa 4/9/07
Vatia 4/13/07
Aoa 4/20/07
Aunu’u 5/16/07
Alofau 2/5/07
Amaua 3/18/07
Faga’alu 2/16/07
Faga’alu schoo 5/907

6
Matu’u 4/17/07
Faganeanea 4/17/07
Nu’uuli 4/3/07
Fagatele 3/23/07
Leone 2/17/07
Aua 5/8/07
Fagatogo 4/19/07
Utelei 2/6/07
Leloaloa 4/27/07
Onesosopo 5/8/07

Results
Benthic
For background information on the coral reefs of American Samoa, see Wells (1988),
Craig (2005), Craig et al. (2005), Sabater and Tofaeono (2006, 2007), Whalen and Fenner
(2006), Fenner (2008), Fenner et al. (2008), Birkeland et al. (2008), and Brainard et al.
(2008).

Reef Slopes
In addition to the 11 sites done in previous years, one new site was added, Masacre
Bay, to improve geographic balance. Figure 1 presents the results for all 12 sites. As in
the past, sites in this graph are ordered from left to right beginning with Fagamalo on the
northwest corner of the island, and proceeding clockwise around the island. So the sites
from Fagamalo to Aoa on the left of the graph are on the north side, and sites from
Amaua to Leone on the right side are on the south side of the island.

7
Figure 1.

Benthic Cover, 2007

zoanthid
100% ascidian
90% cyanobacteria
80% sponge
Percent Cover

70% soft coral

60% dead coral with algae
50% macroalgae
40% silt
30% rubble
20% corallimorph
10% sand
0% rock
turf algae
a
o

e
tia

lu

n
u
as
al

el
nu

ea
aa
Va
am

at
g

branching coralline algae

m
Au

g
Fa

g
Fa
g

Fa
Fa

crustose calcarious algae

live coral cover

Figure 2 below re-graphs the data presented in Figure 1, to facilitate comparisons

between sites for the same cover variable, and adds standard error error bars to allow the
reader to determine which differences are significant. If the error whiskers of two bars do
not overlap, the difference is significant, if they do overlap the difference is not
significant.

Figure 2.

Benthic Cover, 2007

Fagamalo
80 Masacre Bay
70 Fagasa
60 Tafau
Percent Cover

50 Vatia
40 Aoa
30 Aunuu
20 Amaua
10 Fagaalu
Nuuuli
0
live coral CCA BCA turf algae corallimorph macroalgae Fagatele
Leone

8
There are many differences in Figure 2 that are significant, as well as others that are not
significant. In Figure 1 it is obvious that there are many differences between sites, and
Figure 2 allows one to determine which of these differences are significant. Many are,
which means that there is a great deal of real spatial variation in benthic cover around
Tutuila.
The new site, Masacre Bay, has relatively high crustose calcareous algae cover, more
typical of south side sites than north side sites. In spite of this, the averages for the north
and south sides were very similar to that in previous years, with more crustose calcareous
algae on the south side, and more turf algae on the north side. Also, fleshy macroalgae
were more common on the north than south, and branching coralline algae were more
common on the south than on the north, as seen in Figure 3.

Figure 3.
ascidian
North vs South
cyanobacteria

100% sponge

soft coral

dead coral with algae

80%
macroalgae
Mean Percent Cover

silt
60%
rubble

corallimorph
40% sand

rock

20% turf algae

branching coralline
algae
crustose calcarious
0% algae
North South live coral cover

The differences in crustose calcareous algal cover and turf on the north and south sides
are shown in Figure 4. As stated before, this difference may be associated with
differences in wave energy on the north and south. For over half the year, wind and
waves come steadily from the east, so that wave action is greater on the south side due to
the angle of the island. For the shorter season roughly January to April, wind and waves

9
are more variable and the north side may receive as much or more wave energy than the
south, though wave action during this season may also be less. Crustose coralline algae
have been said to do best when their surfaces are kept clean of sediment and other algae.
However, more recent information indicates that some crustose coralline algae are likely
to be found underneath turfs as well as under macroalgae such as Halimeda. For this
reason, the difference seen in Figure 3 should be referred to as a difference in visible
crustose calcareous algae, since there may be more that is hidden and not recorded in
these surveys.

Figure 4.

North vs South, Algae

60

50
Mean Percent Cover

40
North
30
South
20

10

0
Turf Visible Crustose Calcareous

This is the third year of the Territorial Monitoring Program of American Samoa.
Mean benthic cover for the 11 core sites that were repeated each of the three years is
presented in Figure 4. As seen in Figure 5, the changes over the three years of the
mean cover are remarkably small. Mean coral cover varied by less than 0.4%, and no
trend is apparent, with less than 0.1% change between 2005 and 2007. The largest
differences of any benthic category were less than 3%. Perhaps the closest to a trend
would be in fleshy macroalgae, which increased slightly from 2005 to 2006.

10
Figure 5.
cyanobacteria
Tutuila Mean Benthic Cover sponge
soft coral
100%
dead coral with algae
80% macroalgae
silt
Percent Cover

60% rubble
corallimorph
40% sand
rock
20%
turf algae
branching coralline algae
0%
2005 2006 2007 crustose calcarious algae
live coral cover
Figure 6 is a re-graphing of the same data, presented in a different format, with
standard error bars added. Only the most abundant biological categories are included.

Figure 6.

Benthic Cover Trends

50

40
Percent Cover

30
2005
20 2006
2007
10

0
ria
ac e
A

so CA
l

rf A

llim d

le

ro ilt
ru h

e
e
ck

cy sp ral

-10
ra

ob ng
r a an
ga

ga
C
BC

ac s
bb
ro

te
co

or

co
C

an o
s
al

al
e

ft
liv

tu

m
co

The standard error bars allow one to examine the data for significant differences- if the
bars overlap the difference is not significant, if they do not overlap the difference is
significant. Figure 5 above shows that changes over time were rarely significant.
Changes in coral cover, crustose calcareous algae, branching coralline algae,
corallimorph, fleshy macroalgae, soft coral, sponge, and cyanobacteria were not
significant. Rock and rubble were significantly higher in the first year than later, but this
is likely due to a shift in recording to recording the turf cover on the rock and rubble.

11
Dead coral with algae had a significant increase, but this was likely due to more careful
recording.
One view of coral reefs is that some disturbance events, such as hurricanes, are
natural and that coral reefs have adapted to them over millions of years. If a hurricane
hits Tutuila an average of about once in 5 years, 300,000 hurricanes would have hit it in
the 1.5 million years the island has been in existence. This view requires that coral reefs
recover from disturbances. The natural state of affairs might be that after each
disturbance kills coral, the corals recover over the following years or decades, until a new
disturbance comes along. Thus reefs would repeatedly cycle between disturbance and
recovery. The reefs of Tutuila have been exposed to a series of disturbances in recent
decades, with a crown-of-thorns starfish outbreak in 1978 that killed most of the coral,
hurricanes in 1981, 1987, 1990, 1991, 2004 and 2005, and mass coral bleaching in 1994,
2002 and 2003. Craig et al. (2005) interpreted the ups and downs of coral cover recorded
in the long-term monitoring programs by Birkeland and Green as being due to periodic
disturbance and recovery. Coral cover was estimated to be about 63% before the 1978
starfish outbreak (Wass, 1982), but is only 28% now. This view suggests that reefs on
Tutuila are likely to be in a recovery stage. On the Great Barrier Reef, individual reefs
that have been damaged by storms show strong recovery over about a decade, with
increases of coral cover of about 5% a year (Emslie et al. in press). But Tutuila shows no
sign presently of increasing coral cover. Rapid recovery from disturbance events, such as
those shown in Australia, are considered signs that the reefs are resilient, that is, they
bounce back after destructive events. Although coral cover in Tutuila is much better than
after the crown-of-thorns outbreak, it is not nearly as high as that estimated before the
outbreak. It is not clear why coral cover is not presently increasing. Perhaps reefs here
are not as resilient as some have thought.
As reported in the 2006 report for this program (Fenner 2008), an analysis of data
from a coral disease program by Dr. Greta Aeby had found a decrease over time in the
percent cover of crustose calcareous algae in Tutuila. The data in Figure 4 do not reveal
such a trend in the present data. This does not appear to be a problem.
The Secretariat of the South Pacific (SPC) PROCFISH program devised a “live
coral index” that consists of the live coral cover divided by all cover, alive plus dead.
Figure 5 shows the trend in this index over the three years, plus the average for the South
Pacific countries included in the PROCFISH program. Figure 7 shows a slight
downward trend in the live coral index, but even the lowest value is much higher than the

12
Figure 7.
Live Coral Index

120

100
Percent Live Coral

80

60

40

20

0
05 06 07 S. Pacific
Average

South Pacific average. The slight decrease over the three years may be a result of a
lowering threshold for the definition of dead coral. All of the coral recorded as dead are
long dead, partly eroded and covered with crustose calcareous algae and/or turf algae.
Deciding what is reef and what is long dead coral is not always simple, as branches
slowly erode down to shorter stumps and begin to blend in with the rough features of the
reef. Since none of the recorded dead corals are newly dead, there appears to be no cause
for concern over the slight decrease in the recorded index. The index, however, should
be a sensitive measure of the loss of living corals, and should be more sensitive than just
the amount of live coral cover.
Gomez et al. (1994a,b) used a mortality index to measure the amount of recently
killed coral on a reef. The index is a ratio of dead coral to dead plus live coral. Thus, it
is exactly the additive inverse of the Live Coral Index used by PROCFISH. The
mortality index has been used to distinguish reefs in Indonesia subjected to pollution
from reefs not subjected to pollution (Edinger et al. 1998). Reefs in that study subjected
to pollution had a mean index of 0.48 (so 48% of all corals were dead), while reference
reefs had an index of 0.25 (= 0.75 live coral index, lower than here). This index has also
been used to distinguish reefs subjected to anthropogenic sediment runoff from reefs
subject to natural sediment runoff (van Woesik and Done, 1997). Thus it appears to be a
useful index of non-point pollution (Risk et al. 2001).
Gomez et al. (l994a) were well aware of the limitations of using live coral cover as
an index of reef health. They state “Despite its wide use, however, coral cover is not
always a reliable measure of reef health.” Maragos (1997) states that “ it seems that coral
cover is only useful as an indicator of reef health when it is monitored at the same sites
over time in habitats where coral cover is expected to be high (in the absence of stress).”
That is because different reefs have different “natural” amounts of coral cover, for a
variety of reasons including the amount of suitable substrate available. They don’t state
it explicitly, but a “live coral index” or “coral mortality index” largely overcomes this
problem, by reporting what proportion of all coral is dead. If an area has low natural
coral cover, but all coral recorded is alive, then the coral community is relatively healthy.

13
However, if a reef has high coral cover but much of it is dead, then the coral community
is relatively unhealthy. Gomez et al. (1994a) reported that 48% of 844 sites in the
Philippines have at least half of the coral cover already dead (coral mortality index of at
least 50%). Although they do not give a mean coral mortality index for the sites, this
indicates that reefs there were on the average much less healthy than they are currently in
American Samoa.
One of the major problems in interpreting such data or comparing different studies is
the criteria for “dead coral.” When coral dies, a process starts in which it is partly
eroded, partly reduced to rubble or sand, and partly consolidated into the reef. Dead
corals become covered with filamentous algae within days, may later become covered
with coralline algae, and are bioeroded and consolidated into reef rock over time. Thus,
initially they are very easy to recognize as dead corals, distinct from reef rock, but over
time they slowly transition from dead coral into reef rock. The amount of dead coral that
is recorded heavily depends on what the criterion is for distinguishing dead coral from
reef rock is, and it is not an easy distinction since it is a continuum with no sharp
demarcation. In this monitoring program, when the outline of a coral could be
distinguished it was generally recorded as dead coral, even though it was virtually always
covered with some combination of crustose calcareous algae and turf. It is not possible at
this point to estimate the amount of time the coral has been dead, though coral that is still
near-white has been dead for a very short period of time, weeks or less. It appears that in
areas where crustose calcareous algae dominate, they will dominate dead surfaces within
about 6 months of death. In spite of all this, in American Samoa on reef slopes currently
there are extremely few recently dead (near-white) coral colonies, so few that in most
years not even a single colony was counted. There are some dead corals where the
outline of the colony is clear but it is covered with crustose calcareous algae and/or turf
algae, and there are relatively few such corals that are badly enough eroded to be
intermediate between recognizable coral and reef rock (that is virtually always covered
with crustose calcareous algae and/or turf). Thus, the figures for live coral index for the
second and subsequent years of the program are likely to be pretty accurate. It is much
less certain whether figures from other locations are comparable, since the criteria for
recognizing dead coral are not reported, and it is not reported whether there are many
colonies that are partly eroded and intermediate between well-preserved dead coral and
reef rock. The situation in backreef pools and reef flats is likely to be different;
approximately half of all the staghorn coral in some of the backreef pools is dead
standing coral covered with turf and/or crustose calcareous algae, and there are patches of
collapsed rubble (which is older, likely over 5 years old) from staghorn as well. There is
little dead coral of other species in the pools.
Another limitation to the live coral index is that when a disturbance event occurs, the
live coral index will show it clearly due to the dead coral, but with time the dead coral
will become much less obvious and at some point will no longer be recorded as dead
coral. If live coral doesn’t recover during this period, the index would indicate recovery
when no such recovery had occurred. This can be seen easily by examining the live coral
cover. On the other hand, if the live coral cover recovers rapidly, the reef could return to
normal health even though there is a lot of dead coral present and the live coral index
does not indicate full recovery. So it appears that live coral cover is still needed in
conjunction with the index to interpret the recovery of the reef.

14
 A recent publication (Sandin et al. 2008) reports on four atolls in the northern Line
Islands. These atolls range from Kingman which has no inhabitants and no history of
fishing, to Kiritimati which currently has a population of 5000 which fish actively.
Kingman has a fish community dominated by top predators, Palmyra has fewer top
predators, Tabaueran less, and Kiritimati very few. Jarvis Island is in an area of high
level of ocean productivity that stretches east-west, and Howland and Baker are in the
same high productivity area but not in as high productivity area as Howland and Baker.
American Samoa is in a low productivity area. Kingman has the highest coral cover and
Kiritimati the least, as shown in Figure 8. Figure 8 also show coral cover at Jarvis,
Howland, and Baker, all of which are uninhabited and unfished and in the same general
geographic area (Jarvis is in the Line Is.). The figure shows that all of the unfished atolls
have higher coral cover than all of the fished atolls and islands. Tutuila has higher coral
cover than Kiritimati, Tabuaeran, and Palmyra, though it definitely has more fishing than
Palmyra.

Figure 8.

Coral Cover

60
50
Percent Cover

40
30
20
10
0
n

is

nd

ila
an

er
a

i
ra

at
yr

rv

tu
la
gm

tim
ae

Ba
m

Ja

Tu
ow
l

bu
n

ri
Pa
Ki

Ki

H
Ta

Sandin et al. (2008) also present data on a combination of benthic covers. They
combined coral and crustose calcareous algae (CCA) in a single category. Coral and
CCA appear to represent a measure of a healthy coral reef, as opposed to reefs with high
fleshy macroalgae and turf algae. The reported the levels of the coral + CCA variable for
their study islands and Tutuila are presented in Figure 9. The graph shows a strong
decrease

15
Figure 9.

Percent Cover of CCA + coral

80
70
60
Percent Cover

50
40
30
20
10
0
Kingman Palmyra Tabuaeran Kiritimati Tutuila
mean

in this combined variable from the most pristine reef at Kingman to the most impacted
reef in the Line Islands at Kiritimati. The mean for Tutuila sites in the present data set
are shown by the bar on the right. The mean for Tutuila is almost as high as for the most
pristine reef in the Line Islands. It appears that this may be a useful index of benthic
coral reef health, and if so it suggests that Tutuila benthic health is relatively good. Coral
cover alone suggests that Tutuila benthic health may be intermediate. The 2006 report
(Fenner, 2008) compared Tutuila coral cover to averages for the Pacific and South
Pacific, and found it slightly better than the averages. Also, Tutuila coral cover was
much better than present Caribbean coral cover. But it was not as high has past Pacific or
Caribbean cover, nor as high as estimates of past coral cover in Tutuila. Thus, Tutuila
coral cover seemed to be at an intermediate level, which would be consistent with the
present analysis. However, the new combined variable of coral + CCA suggests Tutuila
may have a relatively healthy benthic community. Perhaps the combined variable gives a
measure of the overall health and resilience, while the coral cover alone indicates where
the reefs are in the recovery process from a series of disturbance events, beginning with
the Crown-of-Thorns outbreak in 1978 and continuing with periodic hurricanes and mass
coral bleaching.
The combined index of coral + CCA for each site, along with a second combined
category, for fleshy macroalgae and turf algae, are presented in Figure 10. Fagatele,
Leone,

16
Figure 10.

Combined Benthic Categories

100
90
80
Percent Cover

70
60 Turf + MA
50
40 CCA + coral
30
20
10
0
Tafeu

Leone
Fagaalu
Nuuuli
Fagamalo

Aunuu

Fagatele

mean
Fagasa

Aoa

Amaua
Masacre

Vatia

and Aunuu have the highest levels of the CCA + coral variable, and Fagasa and Vatia
have the lowest levels. Fagasa and Vatia have the highest levels of the turf + fleshy
macroalgae variable, and Fagatele, Nuuuli, and Amaua have the lowest levels of turf +
fleshy macroalgae. Figure 11 shows that these two variables are strongly correlated, r =
.9044, p < .001. Because the total maximum cover is 100% and together these variables
provide a large proportion of the cover, they are by necessity correlated. Yet a casual
examination of Figure 10 reveals that there is enough cover by other categories that the
correlation need not be so tight. Further, these two variables need not together provide
most of the reef cover. So perhaps the correlation is not a simple result of the 100%
ceiling.

Figure 11.
Correlation of Combined Variables

70

60

50
Turf + MA Cover

40

30

20
R2 = 0.818
10

0
0 20 40 60 80 100
CCA + Coral Cover

17
 Figure 10 also indicates that the northern sites on the left of the graph are different
from the southern sites on the right side of the graph. This can be seen clearly in Figure
12.

Figure 12.

North-South Differences in Combined

Variables

90
80
70
Percent Cover

60
50 CCA+coral
40 Turf+MA
30
20
10
0
North South

Figure 12 shows that there was more CCA + coral on the south side than the north, and
more turf + fleshy macroalgae on the north side than the south. This is largely driven by
the differences in CCA and turf, as seen in Figure 3. Mean coral cover is actually slightly
higher on the north side than the south side as seen in that figure. It should be borne in
mind that these results are for 8 m depth, and patterns may be different at different
depths.

American Samoa EPA has rated different watersheds on Tutuila according to the
density of the human population and watershed size. Watersheds where categorized in
several categories from pristine to most heavily impacted. Human impact might affect
the amount of “good” benthic cover (crustose calcareous algae + coral) versus “bad”
benthic cover (turf + macroalgae). Figure 13 shows the correlation between watershed
rating and the ratio of “good” benthic cover to total “good + bad” (CCA + coral/ (CCA +
coral + turf + MA). The correlation was very low (r = 0.111) and not significant (p < .2).

18
Figure 13
Watershed Rating and Benthic Ratio

1.2
Ratio (high = high CCA + coral)

0.8

0.6
R2 = 0.0125
0.4

0.2

0
0 1 2 3 4
Watershed Rating (0 = pristine)

Bioerosion and indicator species

Bioerosion is when organisms erode the calcium carbonate that makes up the reef.
There are a variety of organisms that cause bioerosion on coral reefs, such as parrotfish,
sea urchins, boring bivalves (clams) and boring sponges. The largest parrotfish,
bumphead parrotfish (Bolbometopon muricatum) erode far more reef than any other
parrotfish, by biting off chunks of live coral and coral rock. They are very rare in
American Samoa. Other parrotfish do some eroding, with Chlorurus microrhinos
causing the second largest amount of bioerosion (Bellwood et al. 2003). They are not
common here. Most sea urchins are uncommon to rare on the reefs of American Samoa,
except in some very specific areas, such as a narrow zone of outer reef flat in Nuuuli.
The one abundant urchin, Echinostrephus molaris, is a small urchin that erodes a
cylindrical burrow about 2 cm diameter that it lives in. Boring bivalves often burrow into
living massive corals, and can be easily seen protruding for some species (e.g., Andara
and Spondyllus) or detected by paired siphon holes in the surface of the coral. They were
not seen and are presumed very rare. Boring sponges can cause large amounts of
bioerosion of reef material. Further, boring sponges are filter feeders that feed on small
organisms and organic material that they filter out of the water (as do the boring
bivalves). Areas with large nutrient inputs from humans such as sewage, such as some
Indonesian reefs and the Florida Keys, have large amounts of boring sponges. Such reefs
are often covered with a “snow” of white silt composed of the microscopic chips of
calcium carbonate excavated and expelled by boring sponges. This white silt is
commonly trapped in the turf algae on rock surfaces. The author photographed such
“snow” in the Florida Keys, and has observed it in the Caribbean and Philippines.
However, boring sponges (Cliona or Siphonodyction) have not been reported from
American Samoa among the 45 species of sponges found in an intensive survey for
introduced species which concentrated work in Pago Pago harbor (Coles et al. 2003).
Pago Pago harbor has turbid green water indicating phytoplankton fueled by nutrient

19
input that is not flushed out of the harbor. The harbor is also reported to have a dense
sponge community in deep water, including some sponges as large as small cars (Paul
Brown, personal communication). Although the harbor would be the most likely place to
find boring sponges, no species of Cliona or Siphonodyction was reported from there
(Coles et al. 2003). The author has not observed or photographed boring sponges
anywhere in American Samoa. Further, reef slopes do not have a coating of the white
“snow” between corals, and the turf algae appears to be clean of such silt.
This evidence appears to be consistent with the view that boring sponges are rare or
absent in American Samoa, and that bioerosion is particularly low in American Samoa.
Further, it is consistent with the observations that many filter feeders are uncommon to
rare in American Samoa, and the reef slopes have relatively clear water. The harbor
shows strong signs of eutrophication, with green water, abundant filter feeding oysters,
and a deep community of large sponges.
Risk et al. (2001) proposes another measure of nutrient input to reefs. This is the
amount of “coral associates.” They define coral associates as organisms living on the
surfaces of corals, such as sponges, polychaetes, bivalves, tunicates and hydroids.
Almost all of these are filter feeders. They are visible on the surface and can easily be
counted on corals such as massive Porites. For instance, in the New Caledonia lagoon,
large boring bivalves are prominent on many massive Porites colonies. However, in
American Samoa, the author has never seen bivalves on massive Porites colonies, nor
sponges, tunicates, and hydroids. The filter feeding polychaete Spirobranchus gigantea
does occur on massive Porites in American Samoa, but is rare. This is consistent with
low nutrient inputs to American Samoa reefs. However, the vermetid mollusk,
Dendropoma gigantea is common. Vermatids live in the surface of corals, and secrete
mucus threads that get spread into nets. The nets catch plankton. However, this mucus
gets draped on the coral surface and probably pulls coral mucus in with it when the
animal pulls the net it to eat it. Thus, it is probably not primarily a filter feeder (Fenner,
in preparation).
When corals die, they are very rapidly colonized by turf algae. Within a few days of
death on the reef flat, corals turn light yellow, then in about a week or two green, then in
a few more weeks, black. Jacobson (2008) reports that when disease kills coral in the
Marshall Islands at Majuro where there is considerable nutrient runoff the dead surface
turns almost black from turf algae. However, away from nutrients, at Arno Atoll, the
surface turns a golden yellow due to the growth of diatoms. This may be a rough assay
of nutrient conditions on a reef. The fact that dead corals on the reef flats turn green and
then black within weeks suggests significant nutrient inputs on the reef flats.
Observations need to be made on outer banks of killed coral surfaces to see if they turn
golden yellow. Outer banks are much farther from the nutrient runoff from land than the
reef flats are, and they are in water that is clearer than on the reef slopes.

Individual Sites
Now that three years of data are available, it is possible to start looking for trends at
individual monitoring sites. We will now look at each of the 11 core sites, one at a time,
starting at Fagamalo in the northwest and moving clockwise around the island as in
Figure 1. Figure 14 presents the trends in Fagamalo.

20
Figure 14.

Fagamalo Benthic Cover

100% dead coral with algae

90% macroalgae
80% silt
70%
Percent Cover

rubble
60%
corallimorph
50%
sand
40%
rock
30%
20% turf algae
10% branching coralline algae
0% crustose calcareous algae
2005 2006 2007 live coral cover

Fagamalo had an increase in coral cover over the three years of 12%, while crustose
calcareous algae decreased by 11%, and turf algae decreased by 5%. Examination of
standard error bars indicates that only the increase in coral cover from 2006 to 2007 is
significant, the other differences are not significant. It should be born in mind that the
sample for an individual site (400 points in 4 transects) is much smaller than for the mean
of all sites for the island (4400 points in 44 transects for the 11 core sites). As a result,
random effects will be much larger for individual sites than for the whole island, and the
ability to detect real changes is much less. Transect tapes cannot be returned to the exact
same location year after year, even by using stakes, as the exact route taken in laying the
tape will vary, and surge often moves the tape back and forth more than the diameter of
most coral colonies, in spite of the best attempts to secure the tape. The result is an
irreducible minimal level of random variation. Some monitoring programs seek to
maximize the randomness of the placement of recording points, by randomizing the
placement of transects, or by the randomizing of point placement on a video belt
recording by programs such as Point Count. We have attempted to keep the random
variation from year to year modest by trying to place transects as close to the same
location as possible year after year.
Benthic cover over the three years at Fagasa are shown in Figure 15.

21
Figure 15.
cyanobacteria
Fagasa Benthic Cover sponge
soft coral
100%
dead coral with algae
90%
macroalgae
80%
silt
Percent Cover

70%
60% rubble
50% corallimorph
40% sand
30% rock
20% turf algae
10% branching coralline algae
0% crustose calcareous algae
2005 2006 2007
live coral cover

Figure 14 shows an increase in coral cover of 10%, a decrease in turf of 12%, a decrease
of crustose calcareous algae of 5%, a decrease of branching coralline algae of 7.5%, a
decrease in rubble of 12%, and an increase in fleshy macroalgae of 21.5%. The increase
in coral from 2005 to 2007 was significant, as was the decrease in crustose calcareous
algae and branching coralline algae. Turf algae was significantly less in 2007 than in the
previous two years, and the increase in fleshy macroalgae from 2005 to 2006 was
significant.
Benthic cover at Tafeu is shown in Figure 16. There was a slight increase in coral

Figure 16.

Tafeu Benthic Cover

dead coral with algae
100% macroalgae
90% silt
80% rubble
Percent Cover

70%
corallimorph
60%
sand
50%
40% rock
30% turf algae
20% branching coralline algae
10% crustose calcareous algae
0%
live coral cover
2005 2006 2007

cover of 6%, a decrease in turf algae of 14%, and an increase in corallimorph of 10%.
The decrease in turf from 2005 to 2007 was significant, but the changes in coral and
corallimorph cover were not. The corallimorph, Rhodactis sp., forms clones that spread to

22
cover more area, and seem to prefer to cover unoccupied substrate. This species is shown
in Figure 17. It is far more common at Tafeu at the survey depth (8 m) than any of the
other sites surveyed in this program. It is eaten traditionally, but must be cooked to
remove toxicity (Madrigal, 1999). It appears that over the three years in Tafeu it has
expanded into areas of turf algae, so although it is expanding, it is not expanding at the
expense of live coral, which actually increased. Rhodactis has adaptations to withstand
high illuminance in shallow areas, and forms patches on reef flats in the Red Sea (Kuguru
et al. 2007). In Tafeu as well as Vatia, it forms large patches on the slope, in shallow to
medium depths. Rhodactis can develop powerful short stinging tentacles along their edge
for attacking hard corals (Langmead and Chadwick-Furman, 1999a). They are able to
grow over zoanthids, hydrozoan corals, sponges, and encrusting macroalgae, and damage
or grow over hard corals with small polyps that are low on the aggressive hierarchy in
hard corals. Contact with corals can induce the development of the short stinging
tentacles on the edge of the large corallimorph polyps. However, Platygyra corals can
attack and severely damage the Rhodactis polyps. Thus, Rhodactis seems to be in the
middle of the hard coral aggression hierarchy (Langmead and Chadwick-Furman, 1999a).
They reproduce asexually by division, and can double in numbers in one year (Chadwick-
Furman and Spegiel, 2000). Work et al. (2008) documented the spread of Rhodactis
howseii around a shipwreck in Palmyra until it dominated a large area. The presence of
the same species around steel buoys at Palmyra supported the view that iron enrichment
may have been responsible. Tafeu appears to be the most pristine of all of the core sites,
surrounded by cliffs and with no people living in its watershed. Further, it has the highest
coral cover of any of the sites. It is present in significant quantities in Vatia Bay as well,
which has a significant human population in the watershed and less flushing than Tafeu.
There may be some natural iron runoff from the island; orange streaks can be viewed on
cliffs running down from cracks where freshwater emerges. Thus, it seems unlikely that
the expansion of the corallimorph at Tafeu is a result of human activities, rather it appears
natural and innocuous at this point, but it should be tracked in future years.

23
Figure 17. The corallimorph, Rhodactis sp. (mata-malu)

Benthic cover at Vatia is presented in Figure 18. Coral cover has decreased by less
than 4% over the three years, and since the decrease has not been monotonic, this may be
random variation in transect placement. While turf has remained relatively constant,
fleshy macroalgae has increased by 15%, while sand has decreased by 10%, though most
of these changes occurred in different years, with the increase in fleshy macroalgae
occurring first, and the decrease in sand second. The increase in macroalge from 2005 to
2006 is significant, as is the decrease in sand from 2006 to 2007. For the first two years,
the macroalgae consisted of Halimeda green calcareous algae, but in 2007, there was a
sudden appearance of a significant amount of Dictyota brown macroalgae. This is shown
in Figure 18. In Figure 18, the fleshy macroalgae category is divided into Halimeda and
Dictyota. Not a single point of Dictyota was recorded in the first two years, and in 2007
it suddenly had 15% cover. It was observed growing over live corals and Halimeda, and
probably over sand as well. Although coral cover decreased over the three years by
nearly 4%, it actually increased between the last two years when the Dictyota appeared
by 5%. From the graph it appears

24
Figure 18.

Vatia Benthic Cover sponge

soft coral
100% dead coral with algae
90% macroalgae
80% silt
70% rubble
Percent Cover

60% corallimorph
50% sand
40% rock
30% turf algae
20% branching coralline algae
crustose calcareous algae
10%
live coral cover
0%
2005 2006 2007

that it may primarily be growing over sand. A photo of Dictyota is shown in Figure 20.

Figure 19.

Vatia Benthic Cover sponge

soft coral
100%
dead coral with algae
90%
Dictyota
80%
Halimeda
Percent Cover

70%
silt
60%
rubble
50%
corallimorph
40%
sand
30%
rock
20%
turf algae
10%
branching coralline algae
0%
crustose calcareous algae
2005 2006 2007
live coral cover

25
Figure 20. The brown macroalga, Dictyota sp.

Dictyota is one of the types of brown algae that have overgrown newly dead corals in
the Caribbean as the phase shift from corals to algae occurred there, so it is a cause for
concern. The cause of the sudden appearance of Dictyota is not immediately obvious.
Some brown algae including Dictyota can be seasonal. The transect in 2007 was taken in
February, in 2006 it was taken at the beginning of December, and in 2005 at the
beginning of April. Mike King reports a bloom of Dictyota in the Nuuuli (Coconut
Point) pool, which he first saw in January, 2007. It is possible that it appears annually
around January which would account for why it was not seen in December 2006, but it
would have to disappear before April for it to not have been seen in April, 2005. The fact
that it has not been reported before suggests that this is its first appearance. The National
Parks monitoring program has apparently recorded it at several sites, apparently
beginning in 2007. If it was found at remote sites which have no people in their
watershed, then it would be unlikely to be due to human activities.
Another possibility is that conditions such as herbivory or nutrients may have
changed in a way that has encouraged or allowed the growth. Dictyota is chemically
defended and tough with cellulose, and so avoided by most fish. Further, Vatia is a
Communty-based MPA. It seems unlikely that populations of herbivorous fish there
have declined recently. About a year ago, Mike King discovered that high active
phosphate detergent was being imported and sold here. Since then the Governor has
issued an executive order banning the import, but apparently stocks are still being sold.
Vatia is a narrow bay with little circulation or flushing. The monitoring site on the west
side shows signs that it is protected from wave action, by the abundance of Porites
cylindrica, as well as by the lack of surge on dives there. Further, the sand between
corals is grey with a large silt component, apparently from land runoff. This is the only
site with significant sand on a slope. If there was much wave action like on the outer
coast, the wave surge would quickly re-suspend the silt, and it would move the sand
downslope and off the reef. On outer reef slopes, wave surge resuspends silt and move

26
sand downslope depositing it on the shelf, which is why sand is rarely if ever seen on
outer reef slopes here. It could be that the increased nitrogen from laundry detergent has
accumulated in the bay’s water and fueled a bloom of Dictyota in Vatia.
During snorkeling on the reef slope on the east side of Vatia Bay in the vicinity of
the survey sight, it became clear that there are many areas with the Rhodactis sp.
corallimorph on the slope above the transect depths, in shallower water. It is comparable
in abundance to the amount in Tafeu, just at a shallower depth. The Rhodactis was
observed growing over living corals. According to Paul Brown (personal comm.) enough
light passes through the corallimorph that the coral does not die. The corallimorph is
very powerfully attached to the substrate underneath it, but only attaches in the center of
each individual polyp. Coral should be examined to determine whether it dies where the
corallimorph attaches.
Benthic cover in Aoa over the three years is shown in Figure 21. Coral cover
decreased by 6% over the three years, crustose calcareous algae decreased 7.5%, turf
increased by 8.5%, and sand increased by 8%. The coral cover change was not

Figure 21.

Aoa Benthic Cover

dead coral with algae
100%
macroalgae

80% silt
rubble
Percent Cover

60% corallimorph
sand
40% rock
turf algae
20% branching coralline algae
crustose calcareous algae
0% live coral cover
2005 2006 2007

significant, crustose calcareous algae was significantly less in 2007, turf was significantly
more in 2007 than 2006 (but not more than 2005), and sand was significantly more in
2007 than 2005.
Benthic cover in Aunu’u over the three years is shown in Figure 22. Coral cover
decreased slightly by less than 4% over the three year period, crustose calcareous algae
increased by 5%, turf increased by 5%, dead coral with algae increased by 5%, and rock
decreased by 13%. It is quite possible that a light green surface that was recorded as rock
in 2005 was recorded as turf in 2006 and 2007. All but the very most recently exposed
surfaces have enough turf growth to have a green color, even if the turf itself is not
obvious, and in later years have been recorded as turf. The changes in coral and CCA
were not significant, but there was a significant increase in turf from 2005 to 2007. Rock

27
decreased significantly from 2005, and dead coral with algae increased significantly from
2005 to 2006.

Figure 22.

Aunuu Benthic Cover

soft coral
100%
dead coral with algae
90%
macroalgae
80%
70% silt
Percent Cover

60% rubble
50% corallimorph
40% sand
30% rock
20% turf algae
10%
branching coralline algae
0%
crustose calcareous algae
2005 2006 2007
live coral cover

Benthic cover in Amaua over the three years is presented in Figure 23. All changes

Figure 23.

Amaua Benthic Cover

100% soft coral

dead coral with algae
80% macroalgae
silt
Percent Cover

60% rubble
corallimorph
40% sand
rock
20% turf algae
branching coralline algae
0% crustose calcareous algae
2005 2006 2007 live coral cover

were very small and due to random variation.

Benthic cover over the three years at Fagaalu are presented in Figure 24. Crustose
calcareous algae increased by 12% over the three years, but there was considerable
annual variation. Turf algae decreased by 15% over the same period, and again there was

28
significant annual variation. Coral cover was very steady. Between 2005 and 2006 there
was a significant increase in CCA and a significant decrease in turf.

Figure 24.
cyanobacteria
Fagaalu Benthic Cover sponge
soft coral
100%
dead coral with algae
90%
80% macroalgae
70% silt
Percent Cover

60% rubble
50% corallimorph
40% sand
30% rock
20%
turf algae
10%
branching coralline algae
0%
2005 2006 2007 crustose calcareous algae
live coral cover
.
Benthic cover over the three years at Nu’uuli are presented in Figure 25. Crustose
calcareous algae decreased by 15% over the three years, while branching coralline algae
increased by 20%, and rock decreased by 9%. The decrease in CCA from 2005 to 2006
was just significant, and the increase in branching coralline algae was significant. The
decrease in rock was significant. It appears that branching coralline algae may be

Figure 25.

Nu'uuli Benthic Cover

soft coral
100%
dead coral with algae
90%
macroalgae
80%
Percentage Cover

70% silt
60% rubble
50% corallimorph
40% sand
30% rock
20% turf algae
10%
branching coralline algae
0%
crustose calcareous algae
2005 2006 2007
live coral cover

expanding over rock and crustose coralline algae. Coral was relatively steady, so it does
not appear to be growing over coral. Branching coralline algae is not known to be a

29
problem. Most of the branching coralline algae at Nu’uuli is a soft feathery red algae,
probably Cheilosporum spectabile, which may not be significantly calcified.
Benthic cover at Fagatele Bay over the three year period is presented in Figure 26.
The coral cover recorded decreased by 14% over the three year period. However, at the
time of writing, the data for 2008 has just been recorded and the coral cover percentage is
nearly that of the first year. It appears that the transect location in 2006 and 2007 may
have been shifted enough to record lower coral cover. Thus, the apparent decrease in
coral does not appear to be a cause for any concern. Crustose calcareous algae increased
by 14%. The increase in CCA was not significant. In 2007, there was 5% cover by an
encrusting colonial ascidian (sea squirt), Diplosoma simile, shown in Figure 27. This
change was significant. This species was reported by William Kiene to have had
increased cover in another area of Fagatele Bay at about this time, but he reported that it
decreased afterwards. This species is known to grow in between the bases of corals, and
the author has noticed that it is fairly common at the reef flat in Utelei as well as
Fagatele. It has been reported to be able to grow over corals and kill them. In the data
for 2008 just collected, this ascidian did not appear in the transects. It does not appear to
be a problem at this time, but will be watched.

Figure 26.

Fagatele Benthic Cover

ascidian
100% cyanobacteria
90% sponge
80% soft coral
dead coral with algae
70%
Percent Cover

macroalgae
60%
silt
50% rubble
40% corallimorph
30% sand
rock
20%
turf algae
10%
branching coralline algae
0% crustose calcareous algae
2005 2006 2007 live coral cover

30
Figure 27. The compound ascidian, Diplosoma simile.

Benthic cover in Leone over the three years is presented in Figure 28. Coral cover
increased by just 1.5%, crustose calcareous algae increased by 6%, turf increased by 7%,
rock decreased by 10% and rubble decreased by 8%. The increase in turf and the
decreases of rock and rubble were significant, but the coral and CCA changes were not.
Probably some rock and rubble were scored as turf after the first year. The increase in
dead coral with algae from 2005 to 2006 was significant, probably because it was being
recorded more carefully.

Figure 28.

Leone Benthic Cover sponge

soft coral
100%
dead coral with algae
90%
macroalgae
80%
70% silt
Percent Cover

60% rubble
50% corallimorph
40% sand
30% rock
20% turf algae
10%
branching coralline algae
0%
crustose calcareous algae
2005 2006 2007
live coral cover

31
Coral Diversity in Transects

Lifeforms

Corals within transects were recorded both by lifeform, and by genus and species
(where possible), as in previous years. The abundance of different lifeforms in the 12
sites is shown in Figure 29.

Figure 29.

Coral Lifeforms by Site, 2007

250
mushroom
200 digitate
Percent Cover

Acro branch
150
foliose
100 table

50 massive
branching
0 submassive
ga li

n
tia

Au a

Fa aua
M alo

F a cre

N lu
Am uu

column
Le l e
e
u
Ta a

Fa uuu

ea
Ao

on
fe
s

te
Va
ga

n
m

ga
a

m
as
ga

encrusting
Fa

At most sites, the most common lifeform was encrusting, though at Nuuuli, column was
the most common lifeform. This is because at Nuuuli the species Porites rus, which has
a column and plate lifeform, was the most common coral. P. rus is always scored as
column in this program, to keep from having to try to guess with each colony whether the
column or the plate portion is the larger part of the colony (a time consuming and
inaccurate process).
The lifeforms in all of the sites combined is presented in Figure 30. Encrusting is the
most common lifeform, followed by column, branching, massive, and table in that order.
This is consistant with the casual observation that encrusting corals are very common on
most of Tutuila reefs.

32
Figure 30.

Lifeforms in all Transects

800

700
Number of Colonies

600

500

400

300

200

100

0
ve

e
g

m
g

le
n

te
ch

s
in

in
m

oo
ita
si

io
an
st

ta
lu

as

l
nc

hr
g
fo
ru

co

br

di
m

us
a
c

ro
en

br

m
Ac

Figure 31 presents trends in the abundance of lifeforms in transects over the three years
of the monitoring program to date. Only the 11 core sites are included in this graph.
Changes have been relatively minor, with slightly less encrusting and slightly more
branching plus columns in 2007 than in previous years.

33
Figure 31.

Mean Number of Colonies per Site Lifeform Trends, Transects

90
80
70
60
50 2005
40 2006
30 2007
20
10
0
-10
Fo g
an le
ve

ra
e
ra iose
g

Ac Ma s

in

M i t at
n
in

ab

oo

po
Ac rop ssi
m

ch
st

l
ig

hr

ille
lu
ng cru

D
po ra
ol

us

M
br
C
En

o
+

ra

po
ro
hi

ro

Ac
a nc
Br

The standard error bars on the graph above show that there were no significant changes
over time.
The same data on trends is presented in a different fashion in Figure 32.

Figure 32.

Trends in Lifeform Cover

30

25
table
Percent Cover

20
foliacious
15 branching
massive
10
encrusting
5

0
2005 2006 2007

Here, the decrease in encrusting coral cover can be seen to be compensated by an

increase in tables and branching corals. This seems unlikely to be a long-term trend.
The number of lifeforms at each of the sites is presented in Figure 33.

34
Figure 33.

Number of Coral Lifeforms in Transects

10
Number of Lifeforms

9
8
7
6
5
4
3
2
1
0
u

ga i
M al o

Le e
sa

Am u

e
Au a
Fa re

u
Fa ua

n
tia

Fa ul
fe

l
nu

on
Ao

al

ea
te
ac

uu
ga

Va

a
Ta
m

ga

m
as

N
ga
Fa

Trends in the number of lifeforms are presented in Figure 34.

Figure 34.
Trends in Number of Lifeforms

7
Mean Number of Lifeforms

0
2005 2006 2007 .

There was a slight increase in the number of lifeforms from 2005 to 2006, but this does
not appear to be a trend.

Generic and Species Diversity

The cover of the different coral genera at each site is presented in Figure 35.

35
Figure 35.

Coral Cover of Coral Genera in Transects, 2007

60
other
50
Pocillopora
Percent Cover

40 Leptastrea
30 Acropora
20 Porites
Pavona
10
Montipora
0
u

ga i
as l o

Le le
sa

e
Au a
A m 'u
Fa re

N u
Fa ua

n
tia

Fa u u l
fe

on
Ao

al

ea
a

te
nu
ac
ga

Va

a
Ta
m

ga

m
ga
M
Fa

Fagamalo, Tafeu, and Aunu’u were dominated by Montipora, while Vatia and Nuuuli
were dominated by Porites and Fagatele was dominated by Acropora. The mean shows
that Montipora had the most cover on average, followed by Porites, in turn followed by
Acropora. There is a lot of variety between sites. Fagatele is the only site dominated by
Acropora, though Leone and Aunu’u have fairly good Acropora populations. Otherwise,
Acropora is rare. Vatia and Nuuuli are dominated by Porites, with all other genera rare.
Tafeu and Fagamalo are dominated by Montipora, and it is the most common genus in
Aunu’u. Acropora also dominates reef crests, particularly on the south side of the island
(as A. nana), co-dominates back reef pools with Porites cylindrica, and dominates some
backreef areas such as the western part of Fagaitua and Anasosopo. There is also an area
heavily dominated by Acropora at the mouth of Vatia bay. At times there have been
large patches of A. aspera dominating reef flats, but when exceptionally low tides occur
they kill this species.
Figure 36 shows that the most common coral genus in transects was Montipora,
followed by Porites, Acropora, and Pavona, in that order. Almost all Montipora is
encrusting in Tutuila.

36
Figure 36.
Coral Genera in Transects, 2007

10
9
8
7
Percent Cover

6
5
4
3
2
1
0

a
a

na
es

r
a
ra

he
or

tre
po
rit

vo
po
tip

ot
as
llo
Po

Pa
ro
on

pt
ci
Ac

Po
M

Le

Figure 37 shows the trends in coral genera over the three years. There was a slight
decrease in the amount of Montipora in 2007, but slight increases in other genera
compensated for this. None of the differences were significant.

Figure 37.

Trends in Cover of Coral Genera in

Transects

30
other
25
Pocillopora
Percent Cover

20 Leptastrea
15 Acropora

10 Porites
Pavona
5
Montipora
0
2005 2006 2007

Figure 38 presents the number of coral genera in transects at each of the sites. Leone
had the highest number of genera, followed by Tafeu, and Fagasa and Fagaalu had the
lowest number of genera.

37
Figure 38.

Number of Genera per Site

18
16
Number of Genera

14
12
10
8
6
4
2
0
u

ga li
as l o

Le le
sa

A m uu

e
Au a
Fa re

N u
Fa u a

n
tia

Fa u u
fe

on
Ao

al

ea
a

te
ac
ga

n
Va

a
Ta
am

ga

m
g
M
Fa

Figure 39 presents trends in the number of coral genera in transects. There was a
fairly minor decrease in the number of coral genera, which was not significant.

Figure 39.

Trends in Coral Genera in

Transects

14
Number of Genera

12
10
8
6
4
2
0
first year second third year .
year

Figure 40 shows the cover of the most common coral species in the transects at each
site. The graph shows that encrusting Montipora dominated Fagamalo, Tafeu, and
Aunu’u. Porites rus dominated Vatia, Nuuuli, and Fagasa. Although Figure 34 showed
that the genus Acropora dominated Fagatele, this graph does not show dominance by any

38
one Acropora species in Fagatele. No other species dominated any sites. Figure 39
presents the trends in coral species in transects. As with the genera, there was a slight
decrease in Montipora encrusting between 2006 and 2007, but little other change. None
of the changes were significant.

Figure 40.

Cover of Coral Species in Transects, 2007

60
other
50 Acropora hyacinthus
Percent Cover

40 Porites massive
30 Porites cylindrica
Pocillopora eydouxi
20
Pocillopora verrucosa
10
Pavona chiriquensis
0 Pavona varians
u

ga i
M alo

Le l e
sa

e
Au a
Am 'u
Fa re

N u
Fa ua

Porites rus
ti a

n
Fa u l
fe

on
Ao

al

ea
te
nu
ac

uu
ga

Va

a
Ta
am

ga

m
as

Montipora encrusting
g
Fa

Encrusting Montipora, which others have considered Montipora grisea, but which the
author has yet to verify, was the most common species, followed by Porites rus and
Pavona varians, in that order, as shown in Figure 41.

Figure 41.
Coral Species in Transects, 2007

10
9
Percent Cover

8
7
6
5
4
3
2
1
0
r
xi
ns

us
a
sa
s
g

ac e

he
er is

ric
ru
tin

es dou

iv
s
ria

th
co

ot
ss
en

nd
us

in
va

ru
ite

ey

po ma
qu

Po cyli
cr

r
na
en

Po

hy
iri

Po ra v

s
r
ch

po
vo

ite
a

ra
rit
or

po
Pa

r
llo
na

Po
tip

ci
llo
vo

ro
on

ci

Ac
Pa

Po
M

39
 Figure 42 shows trends in the coral species in transects. Encrusting Montipora
dominated the coral community each year, with Porites rus the second most common
species, followed by Pavona varians. Changes were minor wit a decrease in encrusting
Montipora in 2007 being the largest change. These changes are not likely to be due to
real changes in the coral cover.

Figure 42.

Trends in Coral Species in Transects

30
other
25 Acropora hyacinthus
Porites massive
Percent Cover

20
Porites cylindrica
15 Pocillopora eydouxi
Pocillopora verrucosa
10
Pavona chiriquensis
5 Pavona varians
Porites rus
0
Montipora encrusting
2005 2006 2007

Figure 43 presents the number of coral species at each site. Leone had the highest

Figure 43.

Number of Species per Site

35
30
Number of Species

25
20
15
10
5
0
u

ga li
as l o

Le le
sa

A m uu

e
Au a
Fa re

N u
Fa u a

n
tia

Fa u u
fe

on
Ao

al

ea
a

te
ac
ga

n
Va

a
Ta
am

ga

m
g
M
Fa

40
number of coral species, followed by Aunuu and Fagatele. Nuuuli had the lowest number
of species, follwed by Fagasa and Vatia.
Figure 44 presents trends in the number of coral species. There was no trend in the
number of species, supporting the view that the apparent downward trend in genera in
Figure 38 is due to random chance.

Figure 44.
Trends in Numbers of Coral Species in Transects

20
18
16
Number of Coral Species

14
12
10
8
6
4
2
0
2005 2006 2007 .

Different sites differ greatly in the amount of coral recorded in transects. It could be
that the number of coral species is higher at sites with high coral cover simply because
there are more coral colonies. The diversity could be the same, but the corals just closer
together. If the number of coral species were calculated for a standard number of
colonies, the number of coral species might be similar between sites. In Figure 43, the
number of coral species per point of coral in the transects was shown. Figure 45 shows

41
Figure 45.

Number of Species per Unit of Cover

0.5
Species/Unit of Cover

0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0

ga li

n
tia

Au oa

F a aua
as o
F a cre

N lu
Am uu

L e le
e
u
Ta a

Fa uu

ea
M al

on
fe
s

te
Va

A
ga

n
m

ga
a

m
ga
Fa

that the number of species per unit of cover is actually not similar between sites. Fagaalu
has the highest number of species per unit of cover, Amaua second, and Masacre Bay
third. Fagaalu, Amaua and Masacre Bay all have low coral cover as can be seen in
Figure 36, but so does Fagasa, but it doesn’t have a high number of species per unit of
cover. In any case, it is clear that it is not true that all sites have similar numbers of coral
species per unit of coral area. Interestingly, although Leone, Fagatele, and Aunuu had
high numbers of coral species, none of them had a high number of species per unit area.
The coral species diversity measure, H’, is presented for sites in Figure 46. Leone,
Fagatele, and Amaua had the highest coral species diversity. Nuuuli had by far the
lowest diversity, because almost all the coral there was one species, Porites rus.

42
Figure 46.

Coral Species Diversity in Transects, H', 2007

2.5
2
H'

1.5
1

0.5
0
u

Fa uli
o

le
sa

uu

e
a
re

u
Fa a
tia
fe
al

on
Ao

al
au

te
ac

uu
ga

n
Va
Ta
m

ga

ga
Au

Le
Am
as

N
ga

Fa
M
Fa

Trends in coral species diversity as measured by H’ are shown in Figure 47.

Figure 47.

Trends in Diversity, H'

2.5

2
H'

1.5

0.5

0
2005 2006 2007 .

There was a very slight increasing trend in diversity, but that was not significant and so
just due to chance.

Reef Flats
Reef flat data was taken from all reef flats corresponding to the core 11 monitoring
sites for the first time in 2007, except Tafeu which cannot be accessed from land easily
(and crossing the crest can be hazardous). These were supplemented by transects taken at
Matu’u and Faganeanea on the south central coast. Coral cover tended to be better near

43
the reef crest, so at least one transect was taken near the reef crest. At some sites, a
transect was also taken back from the reef crest. The transect near the crest was called
the “outer” transect and the transect farther from the crest was called the “inner” transect.
Transect locations were chosen to be near the slope sites, and specified by landmarks,
given in the methods section.
Figure 48 shows reef flat cover on “outer” transects, along with the mean.

Figure 48.

Outer Reef Flat Benthic Cover sand

BCA
Percent Cover

100% MA
80%
60%
Sponge
40% DCA
20%
0%
Tunicate
Rubble
g alu
FA
N ol
ga li
Le le

M e
n
Fa g a i
sa
Fa alo

tia

Al a
Am au

U a
Fa e l e

Fa u u

on
ea
Ao

au

ho

te
ga
Va

of

CCA
m

u
t

sc
ga

Coral
Fa

As can be seen from the graph, the most abundant benthic category was turf algae, with a
mean of nearly 47% cover, followed by coral with 21% cover and coralline algae with
15% cover. Utelei was different from other sites, with 63% coral cover and 29% tunicate
cover. This is probably because Utelei is inside the harbor.
Figure 49 shows benthic cover on the “inner” reef flats.

Figure 49.

DCA
Inner Reef Flat Benthic Cover
Tunicate
Percent Cover

100% Sand
80% BCA
60%
40% MA
20% Sponge
0%
Rubble
FA
M e
Al u

U a

ga a
L e le

n
A m au

ga u
Au tia

at i
M ele

Fa ne

on
nu

au

Fa afa

ea
te
Va

of

CCA
Coral

44
As can be seen from the graph there was more variability between sites for the inner reef
flat than the outer reef flat. Leone was particularly unusual, completely dominated by a
thin grey encrusting sponge. The amount of rubble was particularly variable between
sites.
The mean cover on the outer and inner reef flats is compared in Figure 50. Coral
cover was significantly higher on the outer reef flat than the inner reef flat, turf cover was
significantly higher on outer reef flat than inner, while rubble and sponge cover were
greater on the inner than outer reef flat, but neither was significant. Sand was
significantly higher in the inner reef flat. The difference in mean sponge cover was
driven completely by the high sponge cover at Leone, and so that difference wasn’t
significant.

Figure 50.

Outer Reef Flats Compared to Inner

100%
BCA
90%
DCA
80%
70% tunicate
Percent Cover

60% MA
50% sand
40% sponge
30% rubble
20% Turf
10%
CCA
0%
Coral
Outer Inner

Figure 51 compares benthic cover on the reef flats to cover on the reef slopes. Coral
cover increased strongly with distance from shore, from inner to outer reef flat to reef
slope. CCA also increased, particularly from outer reef flat to slope. Turf increased from
inner to outer reef flat, then decreased sharply on the reef slope. Rubble decreased with
distance from shore, from inner to outer flat to slope. Coral cover was significantly
greater on the slope than both reef flat areas, as was CCA. Turf was significantly less on
the slope than both reef flat areas, as was rubble.

45
Figure 51.

Reef Flat Cover Compared to Slopes sponge

soft coral
100%
dead coral with algae
90%
macroalgae
Mean Percent Cover

80%
70% silt
60% rubble
50%
corallimorph
40%
30% sand
20% rock
10% turf algae
0% branching coralline algae
inner flat outer flat slope at 8.5
crustose calcarious algae
m
live coral cover

There were very little differences in cover on outer reef flats on the north and south
sides, as seen in Figure 52. Turf algae had less cover on the South than the North,
and rubble had more cover on the South than the North. For inner reef flats, there was
data from only one northern site, Vatia, so North-South comparisons would not be
meaningful.

Figure 52.

North vs. South Outer Reef Flats

100%
90% sand
80% BCA
70% MA
Percent Cover

60% Sponge
50% DCA
40% Tunicate
30% Rubble
20% Turf
10% CCA
0% Coral
North mean South mean

In 2004, Zoe Andrews gathered transect data from the benthic communities on the
reef flats and pools of Tutuila, resulting in a master’s thesis (Andrews, 2004). A

46
summary of the results appeared in Fenner et al. (2008). A graph of the results from that
publication are presented in Figure 53.

Figure 53. From Fenner, et al. 2008.

Reef Flats, Andrews 2004

100%

rock
80%
rubble
sand
% Cover

60%
other
sponge
40%
turf algae
macroalgae
20% coralline algae
dead coral with algae
0% dead coral
soft coral
te uli

y
M atia

-A a

fu
C ua

n
Fa N rol
Al ofa u

a
F a oon
l a at

Ba
to itu

ea
A l efa

au

O
au u fl

g a u'u
A

t
on

live coral
Au ga
V

m
m
g
as

le
a
Au
of

In independent Samoa, the only coral reef monitoring is of reef flats in front of
MPAs. The reef flats there are generally different from the reef flats on Tutuila. Some
are much wider, on the order of 0.5-1 km wide, and many are deeper, around 0.5-1 m
deep. In some areas they are sandy or rubble-bottomed, with patchy coral. In Figure 54,
averages are presented for the Andrews data for Tutuila, the present data for outer and
inner reef flats on Tutuila, and data for Upolou, Savai’I, and Manono in independent
Samoa from Samuelu and Solofa (2007).

47
Figure 54.

Reef Flat Comparison other

tunicate
100% rock
90% rubble
80%
Percent Cover

70% sand
60% other
50% sponge
40%
turf algae
30%
20% macroalgae
10% coralline algae
0% dead coral with algae

o
r
r

M i'I
lu
TM e w s

dead coral
ne
T M u te

on
va
po
in

an
o

Sa
dr

soft coral
P
P
An

live coral

Substrates on reef flats are very patchy, and because there is no slope, the location of
transects is not constrained in one dimension, and transects can vary in location in two
dimensions instead of one as on the reef slope. The results heavily depend on the
location of the transects, both the site and the exact location on the reef flat in two
dimensions. As a result, variation is generally high both between sites within a program
on an island, and between programs and islands. In particular, the Territorial Monitoring
Program (TMP, the present study) found lower coral cover on the reef flats of Tutuila
than the Andrews study, and much higher turf algae. This is almost certainly due to the
differences in the locations of the transects. In the Andrews study, some transects were
in backreef pools, which were dredged to provide material for villages, and a portion of
those pools have sandy substrate. Sponge had more cover in the inner transects in the
present study than any of the other locations or studies, simply because Leone was
included, where the inner reef flat is nearly 100% covered with encrusting sponge. For
the sites in independent Samoa, the categories used were somewhat different, with one
category for algae which may include macroalgae, turf, and coralline algae, and one
category for abiotic which would include sand, rubble, and rock. The data for algae and
abiotic for independent Samoa were put in rubble and turf algae, to make them
comparable to the American Samoa data, but they may in fact have included other
categories.

Coral Lifeforms

Figure 55 shows coral lifeforms on the outer reef flats by site. Reef flats are
dominated by encrusting and branching corals. Utelei, which had unusually high coral
cover, was dominated by branching corals, so the mean cover of branching corals was
greater than that of encrusting corals. Six sites were dominated by branching corals, and

48
Figure 55.

Coral Lifeforms, Outer Reef Flat, 2007

70
60 column
Percent Cover

50
digitate
40
submassive
30
table
20
massive
10
foliose
0
branching
F a aga i

ga l i

n
Al a
tia

a
Am au
Fa alo

g alu

Le le
e
N l
F ele
sa

o
Fa uu

ea
Ao

au

on
encrusting

te
ho
Va

of
ga
am

M
U

sc
g
Fa

two sites were dominated by encrusting corals. One site (Leone) was dominated by
foliose corals. The dominant species of coral there was Pavona frondifera, which could
be called foliose or perhaps submassive. Figure 56 shows more of the details in the cover
of the different lifeforms.

Figure 56.
Coral Lifeforms, Outer Reef Flat, 2007

14
12
Percent Cover

10
8
6
4
2
0
g

te
g

se

ve

ve
bl
m
in

tin

ta
si
lio

si
ta
ch

lu
us

gi
as

as
fo

co
an

di
cr

bm
Br

En

su

Figure 57 shows the coral lifeforms on the inner reef flat by sites. There was high
variation between sites.

49
Figure 57.

Lifeforms, Inner Reef Flat

25

20 column
Percent Cover

15 massive
foliose
10 branching
5 encrusting

0
u

le

e
i
au
a
ia

le
fa

on
au

te
t

te
Va

af
o

ga

Le
Am

U
Al

at
Fa
M

Figure 58 compares the lifeforms on the outer and inner reef flats. Outer reef flats had
much more branching and encrusting coral cover than inner reef flats, but otherwise they
were similar.

Figure 58.

Lifeforms, Comparing Outer and Inner Reef

Flat

14
12
Percent Cover

10
8 outer
6 inner
4
2
0
ve
se
g

n
in

in

um
si
lio
t

ch
us

as

l
fo

co
an
cr

m
en

br

The cover of lifeforms on the reef flats was quite different from that on the reef slopes.
Figure 59 compares the lifeform cover on reef flats with that on reef slopes. Encrusting
corals had much less cover on reef flats than on slopes, while branching corals had much
more cover on outer reef flats than on slopes or inner reef flats.

50
Figure 59.

Lifeforms: Slope Compared to Outer and Inner

Reef Flats

16
14
Percent Cover

12
10 slope
8 outer
6 inner
4
2
0
ve

se

ve
g

e
n

bm te
tin

in

bl
m

ta
si

lio

si
ch

ta
lu
us

as

gi

as
fo
co

an

di
cr

m
en

br

su

The coral lifeforms on outer and inner reef flats in American Samoa is compared to
that reported for independent Samoa in Figure 60. The pattern is quite different, with a
higher diversity of lifeforms on reef flats in Samoa, due to more evenness between
different lifeforms. Likely this would be reflected in species as well, though species have
not been reported from Samoa yet.

Figure 60.

Coral Lifeforms compared for American Samoa

and Samoa reef flats
Average Percent Cover

14
12
10 outer
8
inner
6
4 Samoa
2
0
bm ate

ch
fo e
di e
m ing
an n

ro ive
co g

ve

bl

s
br um
ti n

an
lio
si
ta

su git
ch

Ac ass
us

as
l

br
cr
en

Coral Genera

51
 Figure 61 shows the cover of corals on the outer reef flat by site. Acropora is the
most genus that has the most cover, but it only dominates one site. Montipora is the
second most common genus, which also dominates one site. Utelei was unusual in
having a large amount of Pocillopora, while Leone was unusual in being dominated by
Pavona.

Figure 61.

Coral Genera, Outer Reef Flats, 2007 Psammocora

Porites
70 Pocillopora
60 Platygyra
Percent Cover

50 Pavona
40 Montipora
30 Millepora
20 Lobophyllia
10 Leptastrea
0 Goniastrea
Favites
sa

n
Fa ga i

N ol
ga li
Fa alo

tia

g alu

Le l e
Al a

e
Am au

U a
Fa ele

Fa uu

ea
Ao

on
au

Favia
ho

te
ga
Va

of
am

M
sc

Cyphastrea
g
Fa

Acropora

Figure 62 shows coral genera on the inner reef flats. Porites was the dominant genus.

Figure 62.

Inner Reef Flat Genera

25

20
Percent Cover

15 Psammocora
Pavona
10 Porites
5

0
u

le

e
M ei

Fa u
a
ia

n
fa

on
a
au

ea
te
l
t

te
Va

af
o

ga

Le
Am

m
U
Al

at

52
 Figure 63 compares the genera of outer and inner reef flats. Slopes had more
Montipora than the reef flats, and inner reef flats had less of everything except Porites.

Figure 63.

Coral Genera Compared,

Slopes and Outer and Inner Reef Flats

10
Percent Cover

8
slopes
6
outer
4
inner
2
0
na
a

ra

a
ra

s
or

tre
e
po
po

vo
rit
tip

as
Po
llo
ro

Pa
on

pt
Ac

ci
M

Le
Po

Coral Species

Figure 64 shows the cover of different coral species on the outer reef flats. The

Figure 64.

Coral Species, Outer Reef Flats, 2007

A hyacinthus
70
60 Psammocora contigua
Percent Cover

50 Porites rus
40 massive Porites
30 Pocillopora verrucosa
20 Pocillopora damicornis
10 Acropora aspera
0 Porites cylindrica
Pavona frondens
A m au

ga li
Fa alo

Le le
sa

M e
Fa lei
Al a

g alu
a

n
N ol
tia

Fa uu

on
Ao

au

ea
te
ho
te
ga

of
Va
m

Fa g a

u
U

sc

Acropora nana
ga
Fa

encrusting Montipora

encrusting Montipora species had the most cover, but no one species had a majority of
the mean cover. One site was dominated by encrusting Montipora, and one site was

53
dominated by Pavona frondens. Utelei was dominated by Acropora nana and
Pocillopora damicornis.
Figure 65 presents the coral species on the inner reef flat. There were only five
common species on the inner reef flats.

Figure 65

Inner Reef Flat Coral Species

25

20 Psam contig
Percent Cover

15 P. frondens
P. cylindrica
10 mass Porites
5 Por rus

0
u

le

e
M ei

Fa u
a
ia

n
fa

on
a
au

ea
te
l
t

te
Va

af
o

ga

Le
Am

m
U
Al

at

Figure 66 compares the species on the outer and inner reef flats. There were 11
common coral species on the outer reef flat, but only five on the inner reef flat.
Encrusting Montipora is replaced as the most common species on the outer reef flat by
Porites rus on the inner reef flat.

Figure 66

Coral Species Compared, Outer and Inner Reef Flats

4.5 encrust Mont

4 Acropora nana
3.5 Pavona frondens
Percent Cover

3 Porites cylindrica
2.5 Acropora aspera
2 Pocillopora damicornis
1.5 Pocillopora verrucosa
1 massive Porites
0.5 Porites rus
0
Psammocora contigua
Outer Inner
A hyacinthus

54
 Figure 67 compares the cover of coral species for slopes and outer and inner reef flats.

Figure 67.

Coral Species Cover compared between Slopes and Reef Flats

10
9
Percent Cover

8
7 Slopes
6
5 Outer Flat
4
3 Inner Flat
2
1
0

ra s
rit ns

rit por err

Po opo chir

Po a f era
vo as a
ro ra h

m
Po cyl eyd

Ac hya e
v a

Po vo us

m ca

p o en
Pa ora an
Ac opo int
v
na or
Po aria

da
v

si
Pa es r

Ac ites dri

p
llo nd
v o tip

n
ci na

po as
ra

a
in

o
Pa on

r
tM

Po illo

ra
ll

n
es

p
us

ci
r
ro
cr
en

Encrusting Montipora and Pavona varians had more cover on slopes than flats, and
Porites cylindrica, Acropora nana, Acropora aspera, Pavona frondens, and Pocillopora
damicornis had more cover on outer reef flats than slopes. Porites rus had more cover on
inner reef flats than outer flats or slopes.
Thus, it is clear that the coral communities on the reef flats are quite different from
those on the slopes.

Non-Point Pollution

In continuation of the work presented in the 2006 report on the correlation between
various variables and the America Samoa Environmental Protection Agency impact
ratings for different watersheds, the correlation with the live coral index was computed
again. The results are shown in Figure 68. The correlation was r = 0.2121, p < .2, thus it
was not significant.

55
Figure 68.

Live Coral Index vs. Impact Score

120

100
Live Coral Index

80

60
R2 = 0.0474
40

20

0
0 1 2 3 4
Impact Score

Coral Recruitment

Two sets of coral recruits were found and monitored. One was of Pocillopora
damicornis on dead surfaces of Porites cylindrica that were exposed to air and sun just
offshore of the southwest side (park) of Fagaalu Bay. This was the only place which this
recruitment event was seen. The upper tips of P. cylindrica branches had been killed by
low tides a couple of years earlier, and had eroded to a nearly flat smooth surface. Only
patches of this dead surface were present, and only on the highest points of colonies. A
¼ m2 rebar frame was used to measure colony densities in October or November, 2007.
The maximum diameters of colonies were also measured. Small fragments of the P.
cylindrica also survived, and were measured as well. The P. damicornis were clearly
sexual recruits, while the P. cylindrica were clearly asexual recruits from tiny surviving
patches of tissue on the original colonies. Recruits were counted in six placements of the
frame. An average of 28.8 P. cylindrica colonies were counted per square meter, and an
average of 107.2 colonies of P. damicornis were counted per square meter. Sizes of P.
damicornis were recorded in two frames, and the results are shown in Figure 69.

56
Figure 69.

Pocillopora damicornis Recruit Size Distribution,

Fagaalu

8
Number of Colonies

7
6
5 Location 1
4
Location 2
3
2
1
0
5 10 15 20 25 30 35 40 45 50 55
Maximum Dimension, mm

P. damicornis is relatively common in Fagaalu Bay near the recruitment sites. P.

damicornis broods larvae, which it releases through much of the year, on a lunar cycle.
The larvae are likely to be competent to settle soon after release. So this recruitment
event is likely to be from locally produced larvae.
The second set of recruits is of the table coral, Acropora hyacinthus. The recruits can
be seen on many reefs and some reef flats. They appear to be most common near the reef
crest, both on the reef slope and on the reef flat. On the reef flat, they are most common
within about 1 m of the reef crest where there is a sharp break between the reef flat and
slope, as in Fagasa. On the slope they are usually most common in the upper meter or
two of the slope, but small numbers of colonies are deeper. They are more common on
some reefs than other reefs, but almost all are in the same size range. They were most
common at Fagasa among all the areas observed. Two 50 m transect tapes were placed
on the reef crest or outer edge of the reef flat at Fagasa, and the number of colonies
within 0.5 m of both sides of the tape were counted. A total of 41 colonies were counted
within the 50 m2 area of the first belt transect, and 43 in the second. Thus, there was an
average of 0.84 colonies/m2. In areas they were close enough together that if they reach
maturity they will be overlapping and provide near 100% coral cover. There were no
adult colonies in this area and very few on most reef flats. In some places near the reef
crest there may be small numbers of stumps of dead table corals, indicating that some
colonies lived in those areas at some time previously. The maximum base diameter,
height, and table diameter of 36 colonies were measured. Figure 70 shows the
distribution of base diameters that was recorded. The modal diameter was 6 cm.

57
Figure 70.
A. hyacinthus recruit base size, 2007

10
9
8
Number of colonies

7
6
5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 10 11 12
Diam eter, cm /2

The distribution of heights is shown in Figure 71. The modal stem height was 4 cm.

Figure 71.
A. hyacinthus recruit height, 2007

12

10
Number of Colonies

0
1 2 3 4 5 6 7 8
Height, cm

The distribution of table top diameters is shown in Figure 72.

58
Figure 72.
A. hyacintus recruit table diameter, 2007

12

10

8
Number

0
1 2 3 4 5 6 7 8 9
Diameter, cm

The modal table top diameter was 5 cm. Photographs in Wallace (1999) indicate that
these colonies were likely to be about 3 years old at the time of measurement, on
December 26, 2007. So far, 100% of the colonies are still alive. It is likely that future
growth will primarily be in the diameter of the table top. The size distributions appear to
be interpretable as a single peak and thus a single recruitment pulse. Continuing these
measurements on an annual basis should reveal how rapidly these corals grow, and
hopefully show whether additional recruitment events occur. This recruitment event may
increase the number of table corals on the reefs, and may be part of a continuing recovery
process following a long series of disturbance events.

Harbor
In addition to the reef flats at the core monitoring sites, seven sites were done in the
harbor. The same procedure was followed as on the core sites, except where a reef flat
did not exist. At the Fagatogo site, the transect was taken along the shoreline rock wall at
the same depth, about 2 feet. Toward the head of the harbor from the marina on the south
side, rocks do not extend on the shore below low tide, but there is an offshore bank. That
bank is partly sand and partly Halimeda algae, but it has some lumps of rock. The rock
has very low coral cover and so the cover was taken as zero without running a transect
tape. On the north side just toward the head of the harbor from the ASPA parking lot, the
rock wall is similar to that at Fagatogo, but only 2 coral colonies could be found
anywhere. The rocks were covered with a layer of sediment, probably trapped in
filamentous algae. There were a few encrusting sponges but no other benthic life. There
is very little wave action at any of these sites. There is considerable wave action at the
reef flat site just seaward from the cannery, and the outer harbor sites.
Figure 73 shows the major types of benthic cover at the sites studied on the South side
of the harbor.

59
Figure 73.

Harbor South Side

120

100
Percent Cover

80 other
FA
60
coral
40 CCA
20

0
Marina Fagatogo Utelei

Figure 74 shows benthic cover at the sites studied on the North side of the harbor.

Figure 74.

Harbor North Side

120

100
Percent Cover

80 other
FA
60
coral
40 CCA
20

0
ASPA Leloaloa Aua Onesesopo

As can be seen from the graphs, there is a trend of increasing coral cover toward the
mouth of the harbor, which on the north side primarily replaces crustose calcareous algae,
while on the South side it appears to replace turf algae and other things.
Figure 75 shows the coral cover as a function of distance from the head of the harbor,
with the left line for the South side and the right line for the North side. Coral cover is
clearly a strong function of distance from the head of the harbor. When the data is
plotted as a function of the distance from the mouth of the harbor, the result is similar
(Figure 76). In both cases, coral cover at the same location is lower on the north side
than on the south side. The canneries and ship yards are on the north side, and the cargo
port and fuel port are on the south side. The distance from the mouth of the harbor is
clearly a strong variable, with wave action, currents, flushing and water quality greatest

60
near the mouth of the harbor and least at the head of the harbor. On top of that primary
gradient, the north side shows less coral cover than the south side. The canneries
released large amounts of organic effluents on the north side up to 1991. Nutrient levels
in the harbor waters fell quickly after the cannery stopped releasing effluents near the
canneries (Craig et al. 2005, Figure 11.9). The lower coral cover on the north side of the
harbor may have been produced at least partly by the cannery effluent. The corals on the
rock wall at Fagatogo are almost all Pocillopora damicornis, a tough pioneering species,
and few appeared to be adult size. There were a few colonies of other species. It appears
as though this community may be increasing in colony size and cover.

Figure 75.
Coral Cover on Reef Flats in the Harbor by
Distance from Head of Harbor;
left line South side, right line North

70

60
Percent Coral Cover

50

40

30

20

10

0
0 1 2 3 4 5 6
Distance

61
Figure 76.
Coral Cover on Reef Flats in the Harbor by
Distance from Mouth of the Harbor; left line
North side, right line South side

70

60
Percent Coral Cover

50

40

30

20

10

0
0 1 2 3 4 5
Distance from Mouth of Harbor

A small oyster (Sacrostrea cucculata) can be seen attached to hard surfaces around the
low tide line at some locations in the harbor. They can quickly foul any hard surface in
the harbor near the water line, such as a boat hull, with near 100% cover. Oysters are
filter feeders which feed on plankton. Thus, these oysters have the potential to be a
bioindicator of average plankton populations over a period of time. Casual observation
suggested that there might be patterns of oyster abundance within the harbor, so
population abundances were observed at several locations. An attempt to record cover
with a point intercept technique failed, as the tape could not easily be placed and kept
within the narrow band on the large basalt blocks used to line shorelines. Thus, points
frequently missed the oysters that were present. Instead, a scale of estimates was used
with 3 = abundant, 2 = common, 1 = uncommon and 0 when none were sighted. The
results are presented in Figures 77 and 78, plotted by the distance from the head of the
harbor, plotted on the left. Oyster density was highest near the head of the harbor and fell
off towards the mouth of the harbor in a strong gradient.

62
Figure 77.
Oyster Density as a Function of Distance from
the Head of the Harbor, South Side

3.5
3
Abundance Rating.

2.5
2

1.5
1

0.5
0
0 1 2 3 4
Distance, km

Figure 78.
Oyster Density as a Function of distance from
the Head of the Harbor, North Side

3.5

3
Abundance Rating.

2.5

1.5
1
0.5

0
0 1 2 3 4 5 6
Distance, km

Oyster density is likely to be a bioindicator of plankton density. This was examined

by measuring water transparency in the harbor using a Secci disk. The disc was lowered
from the surface until the disc was no longer visible. The results can be seen in Figure
79. Turbidity of the water was highest near the head of the harbor and lowest near the
mouth. Although Turbidity can be caused by a variety of things other than plankton,
such as suspended silt, the fact that the pattern is the same as with the oysters supports the
view that the turbidity reflects plankton densities. Plankton itself is commonly a
bioindicator of nutrient levels, with higher plankton levels where nutrient levels are
higher. Nutrient levels within the harbor dropped suddenly in 1992, right after the
cannery effluent release point was moved to near the harbor mouth for the pipe and
outside the harbor for the ship-borne high level waste (Craig et al. 2005). However, the

63
present data indicates that nutrient levels within the harbor are still higher than outside
the harbor, and are high enough to produce a plankton bloom and dense populations of a
filter feeder. Further, coral on reef flats which is abundant near the harbor mouth is
absent near the head of the harbor. Around the turn of the century, there were living
coral reefs near the head of the harbor, but dredging and filling has destroyed all of the
reefs near the head of the harbor. However, there are areas of rock substrate suitable for
coral settlement, particularly on the north side of the inner harbor, which have no corals.
That these rocks are suitable substrate for corals is demonstrated by the rock wall in
Fagatago, which has a coral community, which appears to be increasing in size and
density.

Figure 79.
Water Clarity in the Harbor

25
Secchi Disk Reading, m

20

15

10

0
0 1 2 3 4 5
Distance, km from Head

In late June and early July, 2007, reports came in of colored water in the harbor.
Investigation revealed dark reddish-brown water at the head of the harbor. A sample
revealed that the colored water contained many swimming dinoflagellates. A sample was
provided to Don Vargo, and Fred Brooks identified it as Ceratium fuscus. “Red Tides”
are caused by dinoflagellates and are often quite toxic to fish or other marine life at
concentrations in which they do not color the water red. This species is not reported to be
toxic, and no dead or abnormal fish were observed, nor were there reports of irritation by
humans. Some observers reported that the color had come and gone repeatedly
previously, and one or two more periods of colored water were observed, always at the
head of the harbor. It is not known why these blooms occurred, and indeed the causes of
such plankton blooms are not well understood. However, they require sufficient nutrients
for their growth. The turbid green water at the head of the harbor indicates that
phytoplankton is abundant there due to high nutrient levels. Mike King has found that in
recent years, high active phosphate laundry detergent has been imported by local stores
from Asia because it is cheaper. He found high levels in streams below people doing
laundry and low levels above. Don Vargo also thought that fertilizers were being used on
the new soccer field at the head of the harbor. Other typical nutrient sources are

64
piggeries. Whatever nutrients wash into the head of the harbor are likely to accumulate
there, because there is so little flushing. Flushing is greatest near the mouth of the
harbor, and decreases with distance inside the harbor, so anything that is washed into the
head of the harbor accumulates there.

Reef flat data was also taken in Fagaalu Bay. During heavy rainstorms, a large
amount of silt laden very brown water is released by Fagaalu Stream into Fagaalu Bay.
During periods of normal wave action, waves coming over the crest pump water inward
along the south (park) side of the bay, and then out the ava (pass) near the north (school)
side of the bay. Thus, the highly turbid water from the stream during storms usually
flows out on the north side. Intuition suggested that the reef flat at the small park near
the hospital intersection on the north side near the stream mouth was likely covered with
a thick layer of fine silt with nothing left alive on the flat. Surprisingly, the reef flat in
that area is a calcareous sand bank, with some brown silt content. Although
predominantly light grey, it has areas of light brown. When lifted into the water, a plume
of brown to black silt is left in the water. But the predominant material is white
calcareous sand. There are a few lumps of rock, which are partly dead, but have some
live corals (massive Porites) on them, which look reasonably healthy. Nearer to the
school, the bottom suddenly turns to rock. At the crest, this rock has considerable coral
cover, and the coral appears healthy, and there are a variety of species. In between the
corals there is lumpy rock, some of which shows signs of having been live coral some
time in the past, but not recently. The rock is covered with brown silt trapped in
filamentous algae. There are also clumps of Halimeda algae, with a dusting of silt.
Inward from the reef flat the bottom is almost completely dead lumps with just a few
corals. Wave action is minor. It appears that sediment has killed a large number of
corals, particularly away from the reef crest. However, there are a surprising number of
healthy looking coral colonies left near the reef crest, perhaps because the wave action
there can remove enough sediment that they can survive.

Bleaching

Bleaching of staghorn corals (Acropora muricata, A. pulchra and A. nobilis) in

backreef pools at the airport and Alofau were begun in late 2003, and continued in 2007.
Figure 80 shows bleaching at the airport backreef pool.

65
Figure 80.

Percent of staghorn colony Bleaching at Airport Backreef Pool, Tutuila, 2004-2007

100
area partly bleached

80

60

40

20

0
28-Jun 14-Jan 1-Aug 17-Feb 5-Sep 24-Mar 10-Oct 28-Apr 14-Nov 1-Jun
2004 2005 2006 2007

Mass coral bleaching of staghorn corals has continued to occur in the backreef pools
every summer. Four summers of mass bleaching have been recorded, and bleaching on
reef slopes was reported in the previous two years, so it has occurred every summer for at
least six summers and quite possibly longer than that.
Staghorn bleaching was also recorded in the backreef pool at Alofau, and Figure 81
presents a graph of the results.

Figure 81.

Staghorn Bleaching at Alofau

120
Percent Bleaching

100
Deep Pool
80
Shallow Pool
60
Reef Flat
40
Reef Slope
20
0
11/27/2004

11/27/2005

11/27/2006

11/27/2007
3/27/2004

7/27/2004

3/27/2005

7/27/2005

3/27/2006

7/27/2006

3/27/2007

7/27/2007

Staghorn bleaching is more intense at Alofau than the airport, particularly in the deep
pool. The staghorn corals in the deep pool barely finish recovering from one year’s
bleaching when they start to bleach in the next year’s bleaching. If bleaching becomes
more intense with global warming, staghorns in the deep pool could remain bleached

66
year-round. There has been little bleaching on the reef flat, where wave surge moves the
water more and pumps new cooler water over the flat from the cooler, deeper ocean. And
there has been essentially no bleaching on the reef slope during the recording period.
However, mass coral bleaching on the reef slopes were reported in 1994, 2002 and 2003.
Additional mass coral bleaching events on the slopes are sure to come in the future.
The staghorn species that bleach in the back reef pools are Acropora muricata
(=formosa), A. nobilis (?=intersepta) and A. pulchra. Other Acropora species are rare in
the pools, but tend to bleach with the staghorns. The pools are dominated by A. muricata
and the finger coral, Porites cylindrica. P. cylindrica has not bleached, except a small
spot in Fagaalu on the northeast side of the bay. The fire coral Millepora dichotoma
bleaches in the airport pool during the same period as the staghorns, and some years M.
exesa bleaches as well. This is the first reported annual summer mass coral bleaching of
a multi-species community.
Bleaching is produced primarily by high water temperatures, exacerbated by high
light levels. Scott Heron of NOAA plotted sea surface temperatures for an area near to
the airport, on the bleaching graph for the airport. The result is shown in Figure 82.

Figure 82.

Bleaching at Airport Backreef Pool, Tutuila, 2003-2007

160 30
140
Percent of staghorn
colony area partly

29
120

SST (degC)
bleached

100 28
80
60 27
40
26
20
0 25
28-Jun 9-Nov 24-Mar 6-Aug
2004 2005 2006 2007

The correlation between the water temperatures and bleaching intensity is striking.
Particularly striking is the notch in temperatures in 2006 that corresponded to a notch in
bleaching at the same time. It appears that coral bleaching follows the water temperature
on virtually a week by week basis.

Invertebrates
As in past years, there were very few non-cryptic, diurnal invertebrates recorded.
Those that were common enough to record are presented in Figure 83. The orange lump

67
sponge, Stylissa sp., was the most common invertebrate recorded, followed by the thin
grey encrusting sponge Dysdea sp., followed by the small burrowing urchin,
Echinostrephus moliaris. Alpheid (snapping) shrimp, as evidenced by their groove-
shaped burrows was next, followed by a small black feather-shaped hydroid, the
compound ascidian, Diplosoma simile, and a small compound didemnid ascidian. Giant
clams were even less common than any of these, with just 0.4 clams/100m2. No crown-
of-thorns starfish, lobsters, or Triton shells were recorded.

Figure 83.

Invertebrates

30

25

20
No./100m2

15

10

0
s

d
s
a

a
sa

ve
u

oi

ni
de

om
ph
is

dr

em
oo
ys
yl

os
tre

hy
gr
St

id
l
ip
os

D
d
e
e

D
ei
g

in
g

an
on

ph
on

n
h

di
Ec
sp

Al
sp

di

ci
ci

as
in

as
ch
ur

Figure 84 shows trends in the invertebrates recorded. The burrowing urchin,

Echinostrephus moliaris, has increased slightly. The orange lump sponge, Stylissa sp.
has increased dramatically, as has the encrusting sponge, Dysidea sp. Dysidea was not
recorded in the first year, even though it was present. It is not entirely clear whether the
recorded increases are real, or just improvements in the recognition of these species.
Stylissa sp. is a rather obvious sponge, but most of the other invertebrates require a search
image to find or recognize, and likely recording accuracy has been improving. The
Alpheid grooves and hydroid were recorded for the first time this year, just because they
were recognized and a search image formed for the first time. Because the branches of
Pocillopora colonies are not searched, hermit crabs, commensal crabs such as Trapezia
sp. and limpets are not recorded, as they are in the NOAA CRED monitoring. Also,
small commensal crabs are common between the branches of digitate Acropora, but
require dedicated searching to find.
Many invertebrates are nocturnal and cryptic during the daytime. The logistic
difficulties and safety issues of diving at night make it prohibitive at this time, but it may
be possible to do a night dive from the shore at Alofau safely. If there are many cryptic

68
nocturnal invertebrates, the daytime monitoring will grossly underestimate their
abundance. However, sessile invertebrates like sponges and ascidians that are counted in
transects are small and rare compared to some other locations such as the Philippines or
Caribbean, so it is not entirely a difference produced by the nocturnal behavior of some
species. On the other hand, the known diversity of invertebrates here is high, for example
800 species of mollusks are presently known (Fenner et al. 2008), though they are rarely
seen in daytime transects.

Figure 84.

Trends in Invertebrates

70

60
Giant clams
50
Didemnum molle
No./100m2

40 Didemnid sea squirts

30 Sponge Dysidea sp.
Sponge Styllissa sp.
20
Echinostrephus molaris
10

0
2005 2006 2007

Water Clarity
Water clarity was measured horizontally by stretching a transect tape straight,
horizontally at 9m deep. Figure 85 shows water clarity at each site at 8 m depth. Fagasa
had much lower visibility than most other sites.

69
Figure 85.

Water Clarity

30
25
Visibility, meters

20
15
10
5
0
u

Fa uli
o

le
sa

uu
a
re

u
Fa a

n
tia
fe
al

Ao

al
au

ea
te
ac

uu
ga

n
Va
Ta
am

ga

ga
Au

Am

m
as

N
Fa
g

M
Fa

Figure 86 presents trends in water quality. Insufficient readings were taken in 2006 to
have a meaningful measure. Readings in 2005 were primarily taken by Secci disk. One
drawback to using a Secci disk is that sometimes the disc could still be seen at the bottom
near the reef, so the reading had to be taken farther out. Visibility probably increases
with distance from shore. In addition, water clarity can change rapidly due to the amount
of runoff. It is probably too early to tell if the decrease in recorded water quality reflects
any real change.

Figure 86.
Water Clarity

30

25
Visibility, meters

20

15

10

0
2005 2006 2007

Sediment is well known to have a damaging effect on coral reefs, by killing corals.
Corals can clean themselves of small amounts of sediment, but when it gets beyond a

70
certain level they die. Different corals may have different thresholds. Turbid water cuts
light and thus energy and the ability to make food, for corals. Figure 87 shows the
correlation between water clarity and coral cover.

Figure 87.
Water Clarity and Coral Cover

60

50
Coral Cover, percent

40
R2 = 0.0754
30

20

10

0
0 10 20 30
Water Clarity, m eters

For these two variables, r = .2746, p > .1 , so this is not a significant correlation. Figure
88 shows the relationship between water clarity and coral + CCA cover. Both coral and

Figure 88.
Water Clarity and Coral + CCA

100
90
80
R2 = 0.6522
Coral + CCA Cover

70
60
50
40
30
20
10
0
0 10 20 30
Water Clarity

CCA may be affected by sediment, in fact CCA may be more sensitive than coral, since it
can’t clean itself. This correlation is strong, with r = .8076, p < .01, so it is significant.
However, the correlation depends on the two points on the lower left, if they are removed
the correlation drops to only r = .2540, p > .1, so it is not significant. The correlation of

71
water quality with CCA alone is r = .5873, p >.05 , so it is intermediate in strength but
also not significant. The strongest and only significant relationship is between water
clarity and coral + CCA. Still, it depends on just two points, so it needs to be confirmed
in future years.

Species
Tentative identifications of three additional species of algae were made. One is a
brown colored foliose alga that makes flexible but brittle thin horizontal plates about 3
inches in diameter, which is very common on some lower reef slopes, particularly on the
southeast coast of Tutuila, below about 10 or 12 m depth. It may be the species
illustrated in Figure 3.6.1d on pages 121 of Brainard et al. (2008), identified there as a
species of the red calcified alga Peyssonnelia. Photos appear to match photos of
Peyssonnelia bornetii in Payri et al. (2000). It is not shown in either Skelton (2003) or
Littler and Litter (2003), but it is recorded in the list of species by Skelton (2003). A
photo of the live alga is presented in Figure 89.

Figure 89.

A second species was observed and collected by Maloy Sabater on one of the outer
banks. It is a brown alga, with irregularly size and shaped flat stipes up to perhaps 20 cm
long, and covered the bottom in one area. It matches a photo in Payri et al. (2000) of a
brown alga, Spatoglossum solierii, which is not previously reported from American
Samoa.
A golden alga (chrysophyte) has been observed periodically, primarily on shallow
rubble beds, such as in Alofau. It is sometimes called the “Golden Noodle Algae,”
Chrysocystis fragilis. Cells are embedded in a gelatinous matrix, and colonies are fragile
and easily disintegrate. It is known from Guam and Hawaii but not known widely
otherwise. This is the first report of it for American Samoa. It is illustrated in Figure 90.

72
Figure 90.

In 2005, Hawaii Undersea Laboratory (HURL) of NOAA visited on a ship carrying

two research submersibles. D. Fenner joined Dr. Dawn Wright for two dives. On the
shelf seaward of Taema Banks, the outer part of the shelf at about 70-90 m had many thin
discs on it. Samples were collected, and they were identified as giant Foraminifera, of
the genus Cycloclypea. These are the largest of all forams, and typically live in deep
water where their fragile skeleton will not be broken by wave surge. However, they must
have light because they host symbiotic algae (Song et al. 1994).

Recommendations
Benthic – The corallimorph, Rhodactis sp. at Tafeu should be watched in future years,
though it does not appear to present a threat to corals. The encrusting ascidian,
Diplosoma simile, should be monitored in Fagatele Bay as well as elsewhere. The spatial
extent of the brown macroalga Dictyota needs to determined at Vatia and elsewhere on
the north shore, and it should be monitored in different seasons to see if it is seasonal.

References
Andrews, Z. 2004. Evaluation of the effects of community-based fisheries management
on coral reef communities of American Samoa. Masters of Science Thesis, University of
Wales, Bango, UK.

Bellwood, D. R., Hoey, A. S. and Choat, J. H. 2003. Limited functional redundancy in

high diversity systems: resilience and ecosystem function on coral reefs. Ecology Letters
6: 281-285

Birkeland, C., Craig, P., Fenner, D., Smith, L. W., Kiene, W. E. and Riegl, B. 2008. Ch.
20: Geologic setting and ecological functioning of coral reefs in American
Samoa. Pages 741-765 in B. Riegl and R. Dodge, Coral Reefs of the USA, Springer.

73
Brainard, R., Asher, J., Gove, J., Helyer, J., Kenyon, J., Mancini, F., Miller, J., Myhre, S.,
Nadon, M., Rooney, J., Schroeder, R., Smith, E., Vargas-Angel, B., Vogt, S., Vroom, P.,
Balwani, S., Ferguson, S., Hoeke, R., Lammers, M., Lundlblad, E., Maragos, J., Moffitt,
R., Timmers, M., and Vetter, O. 2008. Coral reef ecosystem monitoring report for
American Samoa: 2002-2006. NOAA Special Report NMFS PIFSC.

Craig, P. 2005. Natural History Guide to American Samoa, 2nd Edition. National Park
of American Samoa, Dept. Marine & Wildlife Resources, and American Samoa
Community College, Pago Pago. 96 pages.

Craig, P., DiDonato, G., Fenner, D., Hawkins, C. 2005. The state of coral reef
ecosystems of American Samoa. PP. 312-337 in Waddell, J. E. (ed.), The State of Coral
Reef Ecosystems of the United States and Pacific Freely Associated States: 2005. NOAA
Technical Memorandum NOS NCCOS 11. NOAA/NCCOS Center for Coastal
Monitoring and Assessment’s Biogeography Team. Silver Spring, MD. 522 pp

Edinger, E. N., J. Jompa, G. V. Limmon, W. Widjatmoko, and M. J. Risk. 1998. Reef

degradation and coral biodiversity in Indonesia: effects of land-based pollution,
destructive fishing practices and changes over time. Marine Pollution Bulletin 8: 617-
630.

Emslie, M. J., A. J. Cheal, H. Sweatman, and S. Delean. in press. Recovery, succession,

and convergence of coral and reef fish communities following disturbance. Marine
Ecology Progress Series

Fenner, D. 2008. Results of the territorial monitoring program of American Samoa for
2006, benthic portion. Report to the Department of Marine & Wildlife Resources, the
Coral Reef Advisory Group, and NOAA.

Fenner, D., M. Speicher, S. Gulick, G. Aeby, S. Cooper Alleto, B. Carroll, E. DiDonato,

G. DiDonato, V. Farmer, J. Gove, P. Houk, E. Lundblad, M. Nadon, F. Riolo, M.
Sabater, R. Schroeder, E. Smith, C. Tuitle, A. Tagarino, S. Vaitautolu, E. Vaoli, B.
Vargas-Angel, and P. Vroom. 2008. Status of the coral reefs of American Samoa.
pp 307-331. In J.E. Waddell and A.M. Clarke (eds.), The State of Coral Reef Ecosystems
of the United States and Pacific Freely Associated States: 2008. NOAA Technical
Memorandum NOS NCCOS 73. NOAA/ NCCOS Center for Coastal Monitoring and
Assessment’s Biogeography Team. Silver Spring, MD. 569 pp.

Gomez, ED, Alino, PM, Licuanan WRY, and Yap HT. 1994a. Pages 57-76 in:
Wilkinson, CR, Sudara, S, and Chou LM (eds.) Proceedings, Third ASEAN-Australia
Symposium on Living Coastal Resources, Vol. 1: Status Reviews. Chulalongkorn
University, Bangkok Thailand, May 1994. Australian Institute of Marine Science.

Gomez, E. D., Alino, P. M., Yap, H. T., and Licuanan, W. Y. 1994b. A review of the
status of Philippine reefs. Marine Pollution Bulletin 29: 62-68.

74
Jacobson, D. M. 2008. A four year fine-scale study of a coral disease outbreak in the
Marshall Islands. Poster, 11th International Coral Reef Symposium.

Kuguru, B., Winters, G., Beer, S., Santos, S.R., Chadwick, N.E. 2007. Adaptation
strategies of the corallimorpharian Rhodactis rhodostoma to irradiance and temperature.
Marine Biology 151: 1287-1298.

Langmead, O., Chadwick-Furman, N.E. 1999a. Marginal tentacles of the

corallimorpharian Rhodactis rhodostoma. 1. Role in competition for space. Marine
Biology 134: 479-489.

Langmead, O., Chadwick-Furman, N.E. 1999b. Marginal tentacles of the

corallimorpharian Rhodactis rhodostoma. 2. Induced development and long-term effects
on coral competitors. Marine Biology 134: 491-500.

Littler, D. S. and Littler, M. M. 2003. South Pacific Reef Plants, a diver’s guide to the
plant life of South Pacific coral reefs. Offshore Graphics, Inc, Washington, D. C. 331
pp.

Madrigal, L. G. 1999. Field guide of shallow water marine invertebrates of American

Samoa. American Samoa Government, Dept. of Education. 132 pp.

Maragos, J.E. 1997. Coral reef health in the Central Pacific. Pages 3-29 in Grigg, R.W.
and C. Birekland, Status of Coral Reefs in the Pacific. University of Hawaii Sea Grant
College Program.

Payri, C., de N’Yeurt A., and Orempuller, J. 2000. Algae of French Polynesia. Au Vent
des Iles, Tahiti. 320pp.

Risk, M. J., Heikoop, J. M., Edinger, E. M., and Erdmann, M. V. 2001. The assessment
“toolbox”: community-based reef evaluation methods coupled with geochemical
techniques to identify sources of stress. Bulletin of Marine Science 69: 443-458.

Sabater MG and Tofaeono S. 2006. Spatial variation in biomass, abundance, and species
composition of “key reef species” in American Samoa. Revised Edition. Biological
Report Series no 2006-2. Department of Marine and Wildlife Resources, Pago-Pago,
American Samoa 96799.

Sabater MG and Tofaeono S. 2007. Effects of scale and benthic composition on biomass
and trophic group distribution of reef fishes in American Samoa. Pacific Science 61 (4) :
503-520.

Samuelu, J. I., and A. Solofa. 2007. Samoa. Pages 115-146 in Sulu, R. Status of coral
reefs in the Southwest Pacific: 2004. IPS Publications, University of the South Pacific.

75
Sandin, S., A., Smith, J. E., DeMartini, E. E., Dinsdale, E. A., Donner, S. D., Friedlander,
A. M., Konotchick, T., Malay, M., Maragos, J. E., Obura, D., Pantos, O., Paulay, G.,
Richie, M., Rohwer, M., Schroeder, R. E., Walsh, S., Jackson, J. B. C., Knowlton, N.,
Sala, E. 2008. Baselines and degradation of coral reefs in the Northern Line Islands.
PLOS One 3(2): 1-11.

Skelton, P.A. 2003. Seaweeds of American Samoa. Report to the Department of Marine
& Wildlife Resources, American Samoa. 103 pp.

Song, Y., Black, G., Lipps, J. H. 1994. Morphological optimization in the largest living
foraminifera: implications from finite element analysis. Paleobiology 20: 14-26.

Van Woesik, R., and T. Done. 1997. Coral communities and reef growth in the southern
Great Barrier Reef. Coral Reefs 16: 103-115.

Wallace, C. C. 1999a. Staghorn corals of the world, a revision of the genus Acropora.
CSIRO Publ., Collingwood, Australia. 422 pages.

Wass, R.C. 1982. Characterization of inshore Samoan fish communities. Department of

Marine and Wildlife Resources Biological Report Series 6, Government of American
Samoa. Pago Pago, American Samoa. 27 pp.

Wells, S. 1988. Coral Reefs of the World, Vol. 3, Central and Western Pacific. UNEP,
IUCN. 329 pp.

Whaylen, L. and Fenner, D. 2006. Report of 2005 American Samoa coral reef
monitoring program (ASCRMP), expanded edition. Report to the Department of Marine
and Wildlife Resources and Coral Reef Advisory Group, American Samoa. 64 pp.

Work, TM, Aeby GS, Maragos JE. 2008. Phase shift from a corallimorph-dominated
reef associated with a shipwreck on Palmyra atoll. PloS ONE 3(8): 1-5.

76