See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/225862564

Habitat variables determining the occurrence of

large benthic foraminifera in the Berau area
(East Kalimantan...

Article in Coral Reefs May 2006

DOI: 10.1007/s00338-006-0119-4

CITATIONS READS

46 133

1 author:

Willem Renema
Naturalis Biodiversity Center
110 PUBLICATIONS 1,565 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The origin of marine high tropical diversity: an investigation using cheilostome bryozoans
View project

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Willem Renema
letting you access and read them immediately. Retrieved on: 19 November 2016
Coral Reefs (2006) 25: 351359
DOI 10.1007/s00338-006-0119-4

R EP O RT

Willem Renema

Habitat variables determining the occurrence of large benthic

foraminifera in the Berau area (East Kalimantan, Indonesia)

Received: 31 May 2005 / Accepted: 11 April 2006 / Published online: 8 June 2006
Springer-Verlag 2006

Abstract Knowledge of the assemblage composition of LBF on and around reefs include depth, temperature,
large benthic foraminifera (LBF) in relation to environ- salinity, siltation and hydrodynamic energy (Reiss and
mental conditions is needed to interpret fossil records. In Hottinger 1984; Hallock 1988; Hohenegger 1994;
this study the assemblage composition of LBF is de- Hohenegger et al. 1999; Renema and Troelstra 2001).
scribed for a carbonate shelf with a barrier reef system Many of these parameters are interrelated and in addition
and some reefs outside the barrier. In a total of 140 to magnitude, the seasonal variability is also important in
samples, 3435 species of LBF were found. Four clusters, structuring LBF communities (Renema and Troelstra
roughly corresponding to substrate type were identied. 2001). The same, or very similar combinations of abiotic
Several small groups of samples were recognized that parameters have been related to the development of coral
were collected locally inside, on, or outside the barrier. reefs in various geographical regions (Done 1982; Lirman
Microscale environmental gradients within the substrate et al. 2003; Torruco et al. 2003; Syms and Jones 2004).
or in the benthic boundary layer resulted in spatial dif- Large benthic foraminifera are strongly aected by
ferentiation of the microhabitat of each species. exposure to terrestrial runo and nutrients. In the
Spermonde Archipelago (Southwest Sulawesi) increas-
Keywords Large benthic foraminifera Environment ingly oligotrophic conditions result in (1) a shift in
Assamblages East Kalimantan assemblage composition from low diversity, dominated
by species with chloroplast retention, to high diversity,
dominated by species with algal symbiosis; (2) an in-
Introduction crease in maximum depth of occurrence from 810 m in
nearshore to 45 m in oshore reefs; and (3) increased
Large benthic foraminifera (LBF) are valuable indica- habitat fractionation in the soft substrate assemblages
tors of the past and present status of coral reefs (Hallock (Renema and Troelstra 2001; Renema 2002).
2000). Both LBF and reef-building corals harbour The goal of the present study was to assess to what
symbionts and thrive in very similar environmental degree spatial variation in foraminifera populations can
conditions. In the Spermonde Archipelago (o Makas- be explained by variation in abiotic environmental
sar, Southwest Sulawesi), LBF alfa and beta diversity variables, habitat composition, the distance between
co-vary with those of both corals and reef-associated sample sites, or a combination of the above.
taxa (Renema and Troelstra 2001; Becking et al. 2006).
A good understanding of the biotic and abiotic param-
eters determining the distribution of LBF is essential to Materials and methods
relate their occurrence to coral reef condition.
Important environmental characteristics that previ- The Berau estuary lies immediately north of the Man-
ously have been shown to inuence the occurrence of kalihat peninsula (Fig. 1) in East Kalimantan (Indone-
sia). The study area is on the northern margin of the
Communicated by Geology Editor P.K. Swart
carbonate shelf, where coral reefs are found across a
water gradient from uvially inuenced to fully oceanic,
W. Renema separated by a chain of barrier reefs (Fig. 1). The reefs
Nationaal Natuurhistorisch Museum, PO BOX 9517, inside the barrier have a relatively low live coral cover
2300 RA Leiden, The Netherlands
E-mail: [email protected]
and an abundance of lter feeders such as soft corals and
Tel.: +31-71-5687576 sponges. Rubble is often covered by ne (mud to silt
Fax: +31-71-5687666 grain size) calcareous sediment with the spaces between
352

Fig. 1 a Map of the sampling

area showing sample sites; a 40 Baliktaba 10 km
b
km
b regional setting of the 39 land 0 200 400
< 5 m,coral reef
research area; c depth prole of Mankalihat
Peninsula
< 5 m,sand
the research area. Note the 20 m
Borneo 30
vertical exaggeration, and that 30 m
100m
N
the depth around Kakaban and 31
Sula-
wesi
Maratua exceeds the vertical 32
scale of this gure (350900 m) 14
Panjang
47

46 15 20
20
East Kalimantan 34 33 Derawan 16
1
2 17
3
30
37 29 28
4 Maratua
19

11
7 38 o
6 Kakaban 2 10 N
25 27
35
24 Samama
5 26 9 36
ry
Estua
10 12

Berau
23
Sangalaki
13 22

o
50 118 E 0 20 30 40

c
Derawan- Samama Kakaban Maratua
ridge
0
10
20
30
40

inside barrier barrier outside barrier

the rubble lled with a similar ne sediment. The reef The reefs outside the barrier usually have very steep
base is formed of coarse sand, consisting of coral frag- reef slopes that are aected by very high current veloc-
ments mixed with carbonate mud near the reefs, and ities, resulting in a reef morphology dominated by low
mud dominated further from the reefs. colony heights with an absence of rubble. Apart from
The barrier itself has diverse reef types. The reef at is reef rock, suitable substrate for foraminifera is sparse.
broad, shallow and sand-dominated. Even in the most The reef ats of these reefs consist of reef rock covered
exposed places, there is no rampart, which could provide by turf-forming algae. On the southwest side of Mara-
shelter from high hydrodynamic energy. The ocean tua, the algae grade into a sandy inner zone where sea
facing reef slope is dominated by corals with patches grasses (Thalassia, Halidule) grow.
of rubble. At approximately 1215 m depth a valley, Fieldwork was conducted in October 2003. Sample
1040 m wide, of (coarse) carbonate sand is followed stations were selected to cover a wide range of micro-
by a second coral-dominated, slightly elevated ridge. habitats in a nearshoreoshore transect (Fig. 1). Sam-
The reef base is dominated by coarse carbonate sand. ples were collected by SCUBA-diving, and on the reef
Patches of coral rubble are smothered by carbonate sand at by snorkelling. During a dive, samples were taken at
transported o the reef at, especially on the upper reef 23 m depth intervals starting at 3035 m depth (or
slope. Macroalgae grow predominantly on the outside of shallower if the slope did not extend this far). Conse-
the second ridge. quently, reef base assemblages were not sampled on the
The reef morphology on the slope landward of the reefs outside the barrier, as coral cover extended to more
barrier is more similar to the reefs inside the barrier. than 35 m water depth. For each sample a surface area
Coral cover extends deeper, and the sand at the reef base of 100 cm2 (reef at, sand) or 500 cm2 (rubble) was
is coarser (ne to medium sand). The reef slope is collected. Six categories of substratum were recognized:
dominated by stony corals instead of soft corals and (1) rubble, (2) rubble with sand (open spaces between
sponges. The barrier is interrupted by channels, which rubble were lled with ne carbonate sand), (3) rubble
show dierent reef morphology. Due to high current with algae (foraminifera were both attached to rubble
velocities, most sand is winnowed away, resulting in a and to algae growing attached to the rubble), (4) epi-
seascape consisting alternately of topographical highs phytic (most if not all foraminifera were living attached
characterized by large colonies of corals, and valleys to macroalgae or sea grasses), (5) sand, and (6) reef rock
covered by ne rubble overgrown by macroalgae (large, usually at surfaces of dead coral or rock, often
(especially Hypnea obtusa and Halimeda spp.). covered by crustous coralline algae).
 353

Samples were collected in bags and taken to the value is calculated for each cluster, and the indicator
shore, where the foraminifera were detached from the value of a species is the maximum value calculated for
coral rubble and all samples were sun-dried. After siev- any of the clusters. The indicator value of a species is
ing with a 0.5 mm sieve, the foraminifera that were alive calculated using the following formulas:
when collected were separated from empty tests based
Nij
on their symbiont colours. Aij ;
Besides species belonging to the symbiont-bearing Ni
Nummulitidae, Amphisteginidae, Calcarinidae, Penero- Mij
Bij ;
plidae, Soritidae and Alveolinellidae, Celanthus crati- Mj
culatum was counted, because well-preserved tests often IVij Aij  Bij  100:
show colouring. C. craticulatum does not house symbi-
onts but sequesters chloroplasts from algal food sources In which
and utilizes photosynthate produced by those chlorop- Aij = mean abundance of species i in samples
lasts (Reiss and Hottinger 1984; Lee and Lanners 1988; compared to all clusters
Lee and Anderson 1991). Nij = mean number of individuals of species i across
Cluster-analysis (relative Sorensen, exible b, samples of cluster j
b =
0.75) was carried out on a data matrix deter- Ni = sum of the mean numbers of individuals of
mined by samples and the log-transformed [xi = log species i over all clusters
(densityi + 1)] densities of 26 species (Fig. 2). Species Bij = relative frequency of occurrence of species i
observed in three or fewer samples were not included in in cluster j
the analysis. The outcome of the cluster analysis was Mij = number of samples in cluster j where species i
used for indicator-species analysis by the method of is present
Dufrene and Legendre (1997), resulting in an indicator Mj = total number of samples in cluster j.
value for each species, ranging from 0 to 100. The
indicator value for species x is calculated as a product of Indicator value analysis was done rst by assigning only
relative abundance in a cluster (average abundance of a two clusters with a cut-o level of 40, p < 0.05. Species
given species in samples forming a cluster over the that (1) have IV > 50, (2) show an increase in IV in the
average abundance of the species in all samples) and the step in which the concerned cluster is separated from its
relative frequency of species x in that cluster (percentage sister cluster, and (3) have an IV much higher than in
of samples in a cluster in which species x is present). This other clusters, have been included in the cluster diagram
(Fig. 2) as being characteristic for that cluster. Ultimately
indicator species were determined for seven clusters.
Statistical calculations were done using PC-ORD version
Information Remaining (%) 4 (McCune and Meord 1999).
100 75 50 25 0 A principal component analysis was performed using
the same database. The resulting principal axes were
correlated with a set of parameters like minimum dis-
Cluster 1

tance to shore, minimum distance to ocean, minimum

Amphistegina radiata, IV = 72
distance to river, and water depth.
Heterostegina depressa, IV =63
Calcarina spengleri, IV =80

Baculogypsinoides spinosus, IV = 75
Results
cluster 2

Calcarina defrancii, IV = 65
C. mayorii, IV = 59
Samples (140) containing 03,500 living benthic fora-
minifera were collected from 32 sampling sites. Of these,
95 samples containing more than 50 specimens and more
group with C. defrancii than three species were used in the analysis (Table 1). In
Palaeonummulites venosus, IV = 45
total, 3033 species of living LBF were observed. Dis-
tinct morphogroups were identied within the genera
cluster 3

Operculina ammonoides, IV = 55
Dendritina striata, IV = 63
Celanthus craticulatum, IV = 79 Calcarina (six species, see also Renema and Hohenegger
2005), Peneroplis (three to ve species, only one included
Celanthus, Dendritina
dominated in the analysis) and Dendritina (two species). The pene-
cluster 4

roplids showed a high variability between the samples

Calcarina gaudichaudii, IV = 92
Laevipeneroplis malayensis, IV = 79 and a thorough revision of the generic- and species- level
Neorotalia calcar, IV = 78
taxonomy will be required to classify all specimens.
Fig. 2 Cluster analysis showing the relationship among the four Most of the species occurred on the reef at. The only
clusters recognised. Indicator-values of the characteristic species species that could be consistently recognised was
are shown Peneroplis planatus, which also shows two distinct
 354
Table 1 Summary of the species observed in this study

Species N samples Depth (m) Substrate Area Remarks

X Peneroplis spp. 1520 <2 Seagrass (Halidule?) Kakaban and Maratua 23 Species
Turf-forming algae
Peneroplis planatus 45 0.535 Seagrass (Halidule?) Kakaban and Maratua,
Turf-forming algae Reef base around reefs at barrier
Sand
Dendritina striata 12 1330 Sand Reef base, predominantly
at the outside of the barrier
Dendritina ambigua 6 1330 Sand Reef base, predominantly In fewer samples and usually
at the outside of the barrier in lower density than D. striata,
an exception is station Ber20,
where 45 times as many D.
ambigua as D. striata were found
Laevipeneroplis malayensis 13 02 Turf-forming algae, sea Reef at of Maratua, Kakaban and
grass (Enhalus
and Thalassia) below rubble
at reef at
Sorites orbiculus 63 0.530 Turf-forming algae Barrier and outside barrier
Macroalgae
X Amphisorus spec 1 3 1719 Rubble Sangalaki
Amphisorus cf sauronensis 44 0.525 Reef rock, rubble, macro Oceanic side of barrier Maximum depth at
algae, turf-forming algae and outside barrier Maratua and Kakaban 25 m
Maximum depth at Sangalaki
and Derawan 17 m
X Amphisorus spec 2 1 1 Sea grass: Enhalus Maratua Inside the lagoon
X Marginopora spec 1 1 Sea grass: Thalassia Maratua At SW side of island
Parasorites spp 23 0.51 Below rubble at reef at Maratua and Kakaban Ecophenotypic variation within one,
14.530 Sand at reef base Around barrier reefs but maybe two species
Alveolinella quoyii 18 1430 Rubble, sand Barrier Transition from reef slope to reef
base at the outside of reefs in the barrier area
Amphistegina lessonii 108 335 No preference, but Complete area
highest densities at rubble
Amphistegina lobifera 28 116 Algal turf and rubble Barrier and outside barrier Highest densities at shallowest depths (reef ats)
Amphistegina radiata 105 135 Macroalgae and rubble Complete area Extremely rare at reef at, most abundant
at reef slope
Neorotalia calcar 24 124 Turf-forming algae Kakaban, Maratua Highest densities at reef ats
Macroalgae Derawan, Panjang
Calcarina defrancii 24 121 Macroalgae Barrier
Rubble with sand
Calcarina gaimardi 20 135 Turf-forming algae Maratua and Kakaban Only few specimens at >1.5 m depth, while
at the reef at densities of 12 cm-2
have been observed
Calarina gaudichaudii 21 19 Turf-forming algae, seagrass (Enhalus Maratua and Kakaban
and Thalassia) below rubble at reef at
Calcarina hispida 6 01 Turf-forming algae Maratua
Calcarina mayori 52 335 Macroalgae and rubble Scattered over area More numerous on algae than on rubble
Calcarina spengleri 89 335 Rubble Complete area Most numerous Calcarina, slight
Macroalgae preference for rubble over algae
 355
morphologies (a shallow-living form with a thick test,
found at the reef at of Kakaban

Several groups are in need of a taxonomical revision. Those groups that have distinct morphologies, and can be recognised between samples have been left in open nomenclature. Taxa

and a deep-living one with a thin and aring test).

Twenty-ve species occurred in four or more samples
Highest densities at rubble

13 Specimens per sample

and were included in the analysis.

Highest densities of LBF were observed on the reef
at or living attached to macroalgae at the lower part of
One specimen was

inside the barrier

Highest densities

the reef slope. Because of three-dimensional structuring

(up to 5 cm
2)
Very numerous

in these microhabitats LBF can attain densities of

>5 cm
2. These densities compare with the highest
Remarks

densities observed on Cebu (Renema 2002) and the

Spermonde Archipelago (Renema et al. 2001). Density
on rubble was higher than on rubble covered by sand.
The cluster analysis resulted in four clusters (Fig. 2).
The rst two axes of the principal components analysis
explain 80% of the variation in the entire database
outside barrier

outside barrier

(Fig. 3). The rst axis (r2 = 0.55) correlates positively

Complete area

Complete area

Complete area

with depth and negatively with distance from land. The

Barrier and

Barrier and

second axis (r2 = 0.24) separates the group of samples

Derawan

Barrier
barrier
Ber 38

taken at a sandy substrate from the other samples.

Area

Cluster 1
Rubble with sand

Cluster 1 was characterized by the occurrence of

Amphistegina radiata, Heterostegina depressa and
Rubble, sand
Turf-forming

Macroalgae

Macroalgae

Macroalgae

Calcarina spengleri, without frequent accompanying

Substrate

species. The dominant species in this cluster are typical

Rubble
Rubble

Rubble
algae

species for coral-rubble substrate (often overgrown by

Sand

Sand

Sand
Sand
Sand

coralline algae) (Hallock 1984; Hohenegger 1994;

Renema et al. 2001; Renema and Troelstra 2001;
Renema 2003). Most samples in this cluster were low
in diversity and density (Table 2). All samples taken
from rubble inside the barrier group into this cluster,
Depth (m)

together with samples taken from areas with high

1432

1231

2830
1631
2031
0.51

currents or almost vertical reef walls on and outside

135

835

135

the barrier.
Cluster 2
N samples

indicated by X have not been included in the analysis Cluster 2 contained samples taken from calcareous or
109

leathery macroalgae (Galaxaura, Halimeda, Laurencia

35

72

17
49
3

3
9

and Hypnea), rubble overgrown by the same algae, or

rubble with sand. In addition to the species in cluster
1, three species of calcarinids, Baculogypsinoides
spinosus, Calcarina defrancii and Calcarina mayori,
Palaeonummulites venosus
Operculinella cummingii
characterize cluster 2. These three species occurred in

Elphidium craticulatum
Operculina complanata
Heterostegina depressa

high densities on macroalgae, especially on the ocean

Baculogypsinoides

facing slopes of the barrier reefs and in a lagoonal

Calcarina quoyi

operculinoides setting on the SW side of Maratua. Within cluster 2 a

Heterostegina

ammonoides
small group of samples is characterized by the absence

Operculina
spinosus
of B. spinosus and C. mayori, and an abundance of C.
Species

defrancii. These samples originated from rubble with

Table 1 (Contd.)

sand on the upper slope (815 m) of the ocean-facing

reefs on the outside of the barrier. On the lower part
of the slope (1530 m), algae became the dominant
substrate and C. mayori and B. spinous joined the
assemblage.

X
356

Rubble Dendritina spp. are accompanying species in the reef-

Rubble/sand 3 base assemblage dominated by O. ammonoides.
Rubble/Algae The samples dominated by O. ammonoides and
Epiphytic
Sand
P. venosus were taken from coarse to very coarse sand,
Reef rock with a much larger fraction of the sand (525% weight)
consisting of LBF. This assemblage is very comparable
to the reef base assemblage in the Spermonde Archi-
Axis 2

pelago (Renema and Troelstra 2001).

depth The reefs at Sangalaki, Kakaban and Maratua ex-
tended to such a depth that it was not possible to reach
the reef base. The absence of this cluster outside the
1
barrier might thus be more articial than real.
2
4
Cluster 4

Axis 1 Calcarina gaudichaudii was the most characteristic spe-

cies (IV = 92), followed by Laevipeneroplis malayensis
Fig. 3 Principal components analysis ordination of the Berau (IV = 79) and Neorotalia calcar (IV = 78). The sub-
larger benthic foraminifera data. Data plot into four distinct areas, strate type was predominantly epiphytic on either seag-
as shown by the circles. Note the distinct space for the clusters 3
and 4, and the combined clusters 1 and 2. Cluster 1 appears to be a rasses (Enhalus or Halidule) or turf-forming algae
more restricted fauna than cluster 2 (Gelidiella and Jania). Samples taken from reef rock at
the reef crest also fell within this cluster.
Within the reef ats insucient dierences between
Cluster 3
assemblages taken from turf-forming algae, sea grasses
and reef rock were observed to warrant dierentiation.
Cluster 3 mainly contained samples taken from sand.
However, Peneroplis spp. tended to be most abundant in
Characteristic species were Operculina ammonoides,
seagrass beds, with highest densities being found in
C. craticulatum and Dendritina striata. Within the cluster
patches of Halidule surrounded by ne sand. The density
two subgroups could be dierentiated. These diered in
of Amphisorus sp. decreased going from the reef crest to
the relative abundance of the species characteristic for
the inner part of the reef at.
this cluster, one characterized by high densities of
D. striata and C. craticulatum and a second character-
ized by high densities of O. ammonoides and Pala-
eonummulites venosus. Discussion
Inside the barrier the substrate was a mixture of coral
fragments (13 cm) with ne sand. LBF occurred in Both the cluster and the principal component analyses
relatively low density and only contributed 13% weight show the importance of substrate type for the assemblage
to the sediment. Compared with a similar assemblage in composition of larger benthic foraminifera. The main
the Spermonde Archipelago, the occurrence of Dendri- division into four groups is well explained by the sub-
tina spp. in this assemblage is striking. In the Spermonde strate type of the samples. Cluster 1 was predominantly

Table 2 Some parameters of the four clusters dened in Fig. 2

Substrate type Cluster number

1 2 3 4 Totals
Samples per category

Epiphytic 14 13 27
Reef rock 1 2 3
Rubble 22 11 3 36
Rubble/alga 8 1 9
Rubble/sand 2 1 3
Sand 1 1 15 17
Means SD
Depth (m) 19 8 16 6 20 6 1 0.5 15 9
Number of species 6.4 1.2 8.6 1.9 8.2 1.7 9.1 2.6 8.0 2.1

Numbers of samples that exceeded twice that expected when the substrate types were randomly distributed in all clusters are printed in
bold
 357

collected from rubble, 2 from macroalgae, 3 from sand, cluster 1 and 2 samples were found on the reef slope of
and 4 from turf-forming-algae or seagrass. these stations. Inside the barrier cluster 2 was not been
found at all.
Geographic distribution of the clusters In areas with high current velocities, the benthic
cover remains low and reef debris is immediately ex-
Cluster 4, the reef at cluster, showed the most unam- ported, resulting in limited available substrate for LBF,
biguous spatial pattern (Fig. 4). This occurred on reef and limited shelter from the currents. In shallow areas
ats of reefs outside the barrier. The extensive reef ats Amphisorus is the only species to really cope with these
inside and on the barrier (e.g., those of Pulau Panjang conditions, but deeper down the slope hardly any LBF
and Pulau Samama) were covered by seagrasses, but are found.
living C. gaudichaudii were observed only in a few Steep-sided slopes oer a completely dierent kind of
samples. Samples taken in this area contained low den- limitation to LBF. Apart from a dierent light regime,
sities of Amphistegina lobifera, and were excluded from there is often a constant rain of ne sediment. At ridges
the analysis either because the number of specimens was and crevices where rubble is accumulating the spaces
too low, or because there were fewer than four species in between the rubble is often lled with ne sand, pro-
a sample. viding limited space for LBF and changing the nutrient
Reef topography, especially the turf-forming algae at regime. Additionally, in large aquaria H. depressa shows
the reef crest, is probably only partly responsible for this a clear preference for horizontal or oblique surfaces, and
dierentiation. The occurrence of this assemblage in tends to avoid vertical surfaces (personal observation).
other areas is also related to reefs that are very far from Similar restrictions in the occurrence of LBF have been
terrestrial inuence (Renema and Troelstra 2001; observed at Cebu, Philippines (Renema 2002) and Bali,
Wakatobi, personal observations). It can be speculated Indonesia (Renema 2003).
that physical or chemical parameters of the surrounding Also within cluster 3, the dierences between the two
oceanic waters are important parameters in determining subgroups have a spatial component. The Operculina
the spatial distribution of this assemblage. Palaeonummulites dominated subgroup is found on the
Of the two reef slope clusters, cluster 1 was found reef base at the oceanward side of the barrier and in
throughout the research area, but cluster 2 was not channels through the barrier. The CelanthusDendritina
present at sites inside the barrier (Fig. 4). The C. de- subgroup is predominantly found on the reef bases
francii dominated subgroup of cluster 2 was more inside the barrier. The higher hydrodynamic energy, and
abundant on the oceanfacing slope of the reefs at the probably also the dierent nutrient gradients in ner
barrier (Fig. 4). grained sediments are the most likely candidates to
The sites where cluster 1 was found are at the limits of explain these dierences.
occurrence of most species of LBF and only the most
ubiquitous species remain. Inside the barrier limiting Ecological space
parameters are related to sediment stress and nutrient
concentration, while the sites outside the barrier are Cluster 3 and 4, and the combined cluster 1 and 2 oc-
limited by high current velocity or slope topography. In cupy a very distinct multivariate space in the PCA-
the latter case the limitations are often local and both analysis (Fig. 3). In contrast, there is a complete overlap

Fig. 4 ad Geographic
distribution of samples in which d
cluster 14 have been recorded.
Open circles represent sample
stations in which no samples
from the respective cluster were
found
358

between cluster 1 and 2. All samples in cluster 1 group nutrient levels is much less than that of diatom-
together within a small part of the multivariate space of bearing hosts (John Lee, personal communication).
cluster 2. The lower diversity (Table 2), the negative In surface water around the Berau reefs, dissolved
dierentiation from cluster 2 by the absence of high nitrate and phosphate concentrations are low (data
densities of the three species of calcarinid, and the from world ocean database). Continuous exchange
reduced multivariate space, indicates that cluster 1 is a rates might result in a relatively constant nutrient
restricted fauna compared with cluster 2. availability, but in very low concentrations. Re-
Nutrient ux and herbivore activity control the rel- duced exchange rates with seawater result in higher
ative dominance of dierent groups of benthic primary nutrient concentrations in the interstitial space
producers (Littler and Littler 1984; Hatcher 1990). The within the rubble (DElia and Wiebe 1990). The
maintenance of macrophyte assemblages requires at preference of Amphisorus for larger patches of
least periodically high nutrient ux (Hatcher 1990 and rubble and reef rock in the surf zone supports this
references therein). On the upper slope, corals are the hypothesis.
dominant space-occupying group, while macrophytes 2. Hydrodynamic energy. Amphisorus rmly attaches to
become more important on the lower slope. This tran- the substratum using an organic cement. In this way
sition alters the substrate type and related assemblage it can live in high-energy environments without
composition markedly, and is most probably related to a breaking or being ushed away. Calcarinids are also
change in nutrient supply to the area. considered to be tolerant of high water energy
(though less than Amphisorus). Heterostegina de-
pressa and especially A. radiata are less capable of
Habitat structure living in high-energy environments (Hohenegger
1994; Hohenegger et al. 1999).
Interactions between reef biota and their surroundings 3. Light-levels and spectra. The specic microhabitat
may markedly change environmental parameters, of LBF species is at least partly controlled by light
probably mainly in terms of nutrient concentration intensity (Hallock 1987). The occurrence of A.
and the ratio between particulate and dissolved or- radiata and H. depressa deeper in the rubble might
ganic matter. Above the reef oor, corals and sponges be a result of photoinhibition. No data are avail-
lter particulate organic matter from the seawater and able on the relation between depth at which these
take up dissolved nutrients. In areas with low sedi- species occur in the rubble, and water depth.
mentation rates, rubble accumulates in patches be- However, at least in some deeper reef slope con-
tween coral and colonies. The interstitial space is lled ditions H. depressa has been found to occur at the
with water, leaving a large surface of substrate for rubblesediment interface [personal observation,
LBF. Remineralization of organic biomass, diusion Kepulauan Seribu, Spermonde Archipelago, Wa-
advection with the overlying watermass, and the use of katobi (Indonesia)]. The opaque test of Amphisorus
nutrients aect nutrient concentrations in the intersti- is not as transparent as that of hyaline species, and
tial water (Hatcher 1990). The part of the rubble Amphisorus might need very high light levels to get
patch directly exposed to seawater experiences more sucient photic energy to its symbionts.
exposure to light, wave and current energy. Light
intensity and hydrodynamic energy decrease with The amount of tiering is reduced when the interstitial
depth in the rubble. Although little studied, dissolved space in rubble is lled with sand, and tiering does not
and particulate nutrients change in concentration and occur at all in sandy substrate. Cluster 1 mainly contains
composition from near the rubblewater interface to samples with a very restricted dierentiation within the
deep inside the rubble (see e.g., DElia and Wiebe substrate (limited tiering), while cluster 2 shows a larger
1990). Along this gradient LBF species change. Am- dierentiation, resulting in more species and higher
phisorus is limited to the rubblewater interface, and densities.
C. spengleri and H. depressa, and especially A. radiata On the reef at Amphisorus and Laevipeneroplis live
are found as deep as 1015 cm within the rubble. The in the uppermost part of turf-forming algae, while the
occurrence of monospecic Amphisorus assemblages in calcarinids live in the more sheltered central part.
the breaker zone of the reef crest can be interpreted as Nutrient dynamics in this high energy environment, as
a reduction in the subdivision of vertical space (tier- well as nutrient supplies and light requirements and
ing) within the substrate. tolerances of the foraminifera are not yet known, so it is
Three possible mechanisms causing this pattern can dicult to hypothesize on the causes of this pattern.
be hypothesized: Overall, the following conclusions can be drawn from
this study:
1. Nutrient concentration. The species living at the
rubble surface, Amphisorus sp., contains the sym- 1. Large benthic foraminiferal assemblages in the Berau
biotic dinoagellate Symbiodinium, while the other region can be dierentiated into distinct assemblages
species in the assemblage house diatoms. In culture inhabiting reef at, reef slope and reef base. These
the tolerance of dinoagellate-bearing hosts to three habitats roughly coincide with substrate type,
 359

although the reef slope has both rubble and calcare- Hallock P (1984) Distribution of selected species of living algal
ous algae as the main substrate types. symbiont-bearing Foraminifera on two Pacic coral reefs.
J Foramin Res 14:250261
2. The four clusters all have, to some extent, a spatial Hallock P (1987) Fluctuations in the trophic resource continuum: a
component. This spatial component is strongest in factor in the global diversity cycles? Paleoceanography
the reef at cluster, which was only found in samples 2(5):457471
from reefs outside the barrier. In the other clusters Hallock P (1988) Interoceanic dierences in Foraminifera with
symbiotic algae: a result of nutrient supplies? Proc 6th Int Coral
the spatial component is mainly determined by dif- Reef Symp 3:251255
ferences in relative abundance of the characteristic Hallock P (2000) Symbiont-bearing foraminifera: harbingers of
species (cluster 2 and 3), or the absence of species global change. Micropaleontology 46(Suppl 1):95104
(cluster 1). Hatcher BG (1990) Coral primary productivity: a hierarchy of
3. The reef slope cluster 1 is impoverished compared to pattern and process. Trends Ecol Evol 5(5):149155
Hohenegger J (1994) Distribution of living larger Foraminifera
cluster 2. Although cluster 1 has been found over the NW of Sesoko-Jima, Okinawa, Japan. PSZNI Mar Ecol
entire area, the restricting parameters inside the 15:291334
barrier (terrestrial inuence) and outside (high Hohenegger J, Yordanova E, Nakano Y, Tatzreiter F (1999)
hydrodynamic energy or reef topography) are dier- Habitats of larger Foraminifera on the upper reef slope of
Sesoko Island, Okinawa, Japan. Mar Micropaleontol 36(2
ent. 3):109168
4. Microscale environmental gradients within rubble Lee JJ, Anderson OR (1991) Symbiosis in Foraminifera. In: Lee JJ,
and the benthic boundary layer formed by the reef Anderson OR (eds) Biology of Foraminifera. Academic, London,
biota have a strong eect on the spatial distribution pp 157220
Lee JJ, Lanners E (1988) The retention of chloroplasts by the
of LBF on the reef. foraminifer Elphidium crispum. Symbiosis 5(12):4559
5. Both macro- and microscale gradients in environ- Lirman D, Orlando B, Macia S, Manzello D, Kaufman L, Biber P,
mental parameters appear to be important in deter- Jones T (2003) Coral communities of Biscayne Bay, Florida and
mining the assemblage composition of LBF. The adjacent oshore areas: diversity, abundance, distribution, and
inuence can either be direct, acting at the species (or environmental correlates. Aquatic conservation. Mar Fresh-
water Res 13:121135
species-group) level, or indirect by aecting the Littler MM, Littler DS (1984) Models of tropical reef biogenesis:
substrate type. the contribution of algae. In: Round FE, Chapman DJ
(eds) Progress in phycological research, vol 3. Biopress, Bristol,
Acknowledgements Fieldwork for this study was supported by pp 323364
NWO-WOTRO/KNAW grant WT 75-397. The Indonesian McCune B, Meord MJ (1999). PC-ORD. Multivariate analysis of
Foundation for Research (LIPI) supported the logistics within ecological data. MjM software design, Oregon
Indonesia. I thank Pamela Hallock and an anonymous reviewer for Reiss Z, Hottinger L (1984) The Gulf of Aqaba. Ecological
their valuable comments on the manuscript. micropaleontology. Ecol Stud Anal Synth 50:1354
Renema W (2002) Larger foraminifera as marine environmental
indicators. Scr Geol 124:1263
Renema W (2003) Larger foraminifera on reefs around Bali
References (Indonesia). Zool Verh 345:337366
Renema W, Hohenegger J (2005) On the identity of Calcarina
Becking LE, Cleary DFR, de Voogd NJ, Renema W, Beer M de, spengleri (Gmelin, 1791). J Foramin Res 35(1):1521
Soest RWM van, Hoeksema BW (2006) Beta diversty of trop- Renema W, Troelstra SR (2001) Larger foraminifera distribution
ical marine benthic assemblages in the Spermonde Archipelago, on a mesotrophic carbonate shelf in SW Sulawesi (Indonesia).
Indonesia. Mar Ecol 27:7688 Palaeogr Palaeocl 175(14):125147
DElia CF, Wiebe WJ (1990) Biogeochemical nutrient cycles in Renema W, Hoeksema BW, Hinte JE van (2001) Larger benthic
coral-reef ecosystems. In: Dubinsky Z (ed) Ecosystems of the foraminifera and their distribution patterns on the Spermonde
world 25 coral reefs. Elsevier, Amsterdam Oxford New York shelf, South Sulawesi. Zool Verh 334:115149
Tokyo, pp 4974 Syms C, Jones GP (2004) Habitat structure, disturbance and the
Done TJ (1982) Patterns in the distribution of coral communities composition of sand-dwelling goby assemblages in a coral reef
across the central Great Barrier Reef. Coral Reefs 1:95107 lagoon. Mar Ecol-Prog Ser 268:221230
Dufrene M, Legendre P (1997) Species assemblages and indicator Torruco D, Gonzalez A, Ordaz J (2003) The role of environmental
species: the need for a exible asymmetrical approach. Ecol factors in the lagoon coral community structure of Banco
Monogr 67(3):345366 Chinchoor, Mexico. B Mar Sci 73:2326