Get documents about "Roland
Shared by: keralaguest
-
Stats
- views:
- 7
- posted:
- 2/8/2012
- language:
- English
- pages:
- 23
Document Sample
ENVIRONMENTAL FACTORS INFLUENCING THE RECRUITMENT
OF REEF FISHES
INTRODUCTION
Coral reef fish populations are strongly shaped by the pattern of recruitment they
experience. For many species, the strength of initial recruitment is preserved through
the years, resulting in strong or weak year-classes. Since realizing this connection
(Munro et al. 1973, Williams 1980), there was an increasing research effort to
investigate the dynamics of larval phase and recruitment over several spatial and
temporal scales (see Doherty & Williams 1988 for a review, ), and to identify factors
influencing recruitment and its mechanisms (Vigliola & Meekan 2002).
Most coral reef fish replenish their populations via a larvae with a pelagic phase (Leis
1991) that varies in length from hours to weeks, depending on species and
individuals. Recruitment in the Great Barrier Reef usually occurs between October
and March (Russel et al 1977) and can happen in one or several pulses that often
stretch over broad regions. Pulses are often multi-specific but, generally species
recruit independent from each other and composition and relative abundance of
species in such pulses vary over space and time (Doherty & Williams 1988, Doherty
1996).
There are, however, several factors that can influence a species in their ability to
recruit at all, or at least in their density at recruitment. These can be biological factors
such as density of food or predators, or physical factors like currents, water
temperature (Wilson & Meekan 2002), disturbances or climatic phenomena.
Availability of prey is essential for larval growth while predators have the capacity to
significantly reduce recruitment. Currents may transport larvae away from suitable
habitats, water temperature directly influences larval metabolic rate and growth
(Wilson & Meekan 2002) with faster growing larvae having higher survival rates
(Sirios & Dondson 2000). Moderate disturbances are thought to increase larval
survival by mixing the water and hence bringing them in regular contact with food
items. Major disturbances associated with storms can lead to high mortality in larvae
or newly settled fish by killing them directly, by advecting them away from their
habitat or by destroying their habitat. Finally, global phenomena like El Nino
Southern Oscillation (ENSO) can also possibly influence recruitment by affecting
species directly (see above for the impact of temperature on growth), or by
influencing species that interact with the recruiting fish (e.g. killing corals via coral
bleaching REF).
In this paper we want to primarily investigate recruitment patterns on five reefs in
Great Barrier Reef across the shelf over a period of 18 years. In that we continue
earlier studies that did examine species and recruitment patterns on those five reefs
over the last 20 years (Williams 1982, Done 1982, Williams 1986,...). Secondly, we
investigate how the physical factor temperature and the climatic phenomena cyclones
and ENSO influenced this recruitment.
METHODS
Fish census
Recruitment of reef fishes has been monitored for 2 decades at specific sites in the
central section of the Great Barrier Reef (GBR) adjacent to Townsville. This work is
described in William 1986 and Williams et al 1994 and this report utilizes these data.
Census of the latest recruited juvenile fish were taken on five reefs (Fig. 1). These
reefs constitute a cross-section through the GBR from the nearshore, shallow environs
(Pandora reef) through the midshelf (John Brewer, Lodestone and Rib reefs) to the
outer barrier (Myrmidon reef) abutting the deep oceanic waters of the Coral Sea.
18S
Gillian Feb/97
WS Myrmidon reef
Ita Feb/97
18.5S Rib
John Brewer reef
Lodestone reef
WS
Pandora reef Davies reef
Tessi Apr/00
Ivor Mar/90
19S
Justin Mar/97
Celeste Jan/96
Townsville
Joy Dec/90 Charlie Feb/88
19.5S
146 E 146.5 E 147 E 147.5 E 148 E
Figure 1. Location of study reefs (Pandora, Rib, John Brewer, Lodestone and
Myrmidon) in the Townsville region of the Great Barrier Reef Marine Park, AIMS
weather stations on Davies and Myrmidon and the tracks of cyclones passing through
the region over the study period.
Annual surveys of recruits on each reef were undertaken in late summer/early autumn
over the period from 1983 – 2000 with the exception of Myrmidon in 1997 due to the
presence of a severe tropical cyclone. Three permanent sites per reef were established
on the northern reef slope between 2-7m depth where fish could be found in high
diversity and abundance and the lowest likelihood of weather conditions precluding
survey. Sites were 200-300m apart and consisted of 10/ 5*5m mainly contiguous
quadrats marked by steel stakes and ropes. Their arrangement varied in order to keep
them within the depth range and similar topography and substratum.
The number of recruits of each species was recorded for each quadrat by scuba diver,
one of 2 people who have carried out all censuses. These counts were summed for
each site and then averaged for the 3 sites on a reef.
Environmental Data
Sea Surface Temperature (SST)
Water temperature data was obtained from two sources. The Australian Institute of
Marine Science (AIMS) maintains weather stations on Myrmidon (outer barrier) and
Davies (midshelf south of study sites) reefs. Each station has 2 temperature probes
which integrate continuously and provide SST at 30 minute increments via a
microwave link. Data were available from 1991-2002 and from 1987-2002 at Davies
and Myrmidon, respectively. For various technical and operational reasons, these
records are not continuous but contain frequent gaps where one or both probes
physically failed or returned values outside acceptable limits. Where possible, gaps in
the temporal record were filled from a temperature probe on one of the reefs.
The Integrated Global Ocean Services System (IGOSS)-NMC blended satellite
provided monthly SST of 1° latitude by longitude squares for the period of 1981-
2002. Data from ship, buoy and bias-corrected satellite processed for optimum
interpolation are mingled to yield these SST fields. The Comprehensive Ocean
Atmosphere Data Set (COADS) provided the in situ data for the period from 1981-
1989, and radio messages on the Global Telecommunication System yielded in situ
data from 1990 to present.
The IGOSS SST records for two 1° squares encompassing the surveyed reefs were
chosen and compared for agreement. AIMS SST records were then compared with the
IGOSS data and a continuous monthly SST record established.
El Nino Southern Oscillation (ENSO)
The Commonwealth Bureau of Meteorology (BoM) provided records of the Southern
Oscillation Index (SOI). This index, the differential Tahiti – Darwin sea-level
atmospheric pressure, indicates shifts between El Niño (positive values) and la Nina
(negative values) conditions and the consequent controlling influence over regional
climate. ENSO influence on monthly SST was investigated through SST anomalies
for the IGOSS square 17 for January to December.
The 1998 ENSO event caused massive coral bleaching and subsequent coral
mortality across the GBR. The magnitude of coral mortality on the studied reefs
was investigated using the modeled Future Reef Maps provided by Helix.
Cyclones
Wind speed, barometric pressure, wave height and path for cyclones was available
from the BoM for 1983 to 2000. Cyclones entering the study area (18°S-19°30’S/
146°E-148°E) were extracted from the BoM Geographical Information Systems (GIS)
database.
Daily SST anomalies for a 40 day period centered on a cyclone were contrasted with
the long term anomalies to reveal any impact on temperatures. SST temperature field
charts for the Pacific region were examined for a broader scale interpretation of an el
Niño event.
Some cyclones caused major flooding along the Queensland Coast. These floods and
their resulting impact on nearshore salinity were examined using the Burdekin River
Modeling on the AIMS webpage (http://www.aims.gov.au/pages/research/brpm-crc/).
Habitat
A visual estimate of the percentage of hard coral cover was obtained for each quadrat
at sites on the mid-shelf reefs at the time of the recruitment surveys. Limited data on
coral cover for Pandora reef was provided by Terry Done (AIMS) and the Reef
Monitoring Project (http://www.aims.gov.au/monmap/reefpages/TO/18034S.html) for
Myrmidon reef. Consequently, hard coral cover was specific for sites on the mid-shelf
but relevant only at the reef level for Pandora and Myrmidon reefs.
RESULTS
136 species from 18 families were recorded from the 5 surveyed reefs over the study
period. Recruitment was dominated by Pomacentridae (45 species) and Labridae (38)
with 4 species of pomacentrid accounting for 74% of the overall abundance of all
recruits (Fig 2). 99 fish species had an average abundance on a site of no more than a
single individual (Fig. 3). On average, the mid-shelf reefs had 2 to 3 times more
species recruiting than Pandora or Myrmidon (Fig. 4).
POMACENTRIDAE
LABRIDAE
APOGONIDAE
SCARIDAE
CHAETODONTIDAE
ACANTHURIDAE
CAESIONIDAE
SERRANIDAE
TETRAODONTIDAE
POMACANTHIDAE
PLESIOPIDAE
LUTJANIDAE
BLENNIIDAE
SIGANIDAE
OSTRACIIDAE
MONACANTHIDAE
MICRODESMIDAE
LETHRINIDAE
0 10 20 30 40
Number of species
Figure 2. The number of species by family recorded from annual surveys on the 5
reefs in the central Great Barrier Reef between 1983 and 2000. The unfilled section
indicates the ubiquitous species.
>100
<=100
<=50
<=20
<=10
<=5
<=1
0 20 40 60 80 100
Number of species
Figure 3. The average abundance class of species recruits on a reef demonstrate that
the majority recruit in very small numbers.
outer barrier Myrmidon
Rib
midshelf Lodestone
John Brewer
inshore Pandora
0 10 20 30 40
Average number of species/site
Figure 4. The average number of species per site (3 sites/reef) on each of the 5 reefs
surveyed between 1983 and 2000. Error bar = 1 standard deviation.
18 species of recruiting fishes were recorded from all 5 reefs at some time over the
survey period.
These major differences are also indicated by diversity of new recruits found at the
different reefs (Table7). Total number of species throughout the survey period were
highest for the midshelf (89 species at Rib), followed by Myrmidon (69 species) and
Pandora (55 species). The difference becomes even more striking when looking at
mean species numbers: Rib reef showed 2 times as many recruiting species on
average as in Myrmidon, and almost 3 times as many as at Pandora.
21 species were ubiquitous across the shelf, and 20 to 26% (12 to 18 species) of
species at the crossshelf reefs were unique to these reefs (Table 8). This leaves a high
number of species that were shared only between nearshore and midshelf reef or
midshelf and outer reef (21 respectively 29 species) and suggests shifting patterns of
species recruiting over the shelf.
3. Results:
3.1 Temperature data
Comparing the daily & monthly records of data loggers within one reef showed high
agreement in times when both loggers were active (Table 1). Although one or both
loggers turned inactive for periods at either reef, one of the two loggers always
provided a more continuous record: logger 1 on Davies reef and logger 2 on
Myrmidon. The record of these loggers were chosen in the further comparisons
between loggers of different reefs and loggers and IGOSS data, and whenever a reef
logger was needed in subsequent investigations.
daily Monthly
Davies Logger 1 & 2 0.99 0.99
Myrmidon logger 1 &2 0.99 0.99
Davies1 & Myrmidon2 0.98 0.99
IGOSSsq17 & IGOSSsq18 -- 1
Davies 1 & IGOSSsq17 -- 0.99
Myrmidon 2 & IGOSSsq17 -- 0.98
Table 1: correlation coefficient r
Jan Feb Dec
Average 0.02 0.02 0.04
Difference
Max Difference 0.18 0.12 0.19
Min Difference -0.13 -0.16 -0.08
Table2: IGOSSsq17 – IGOSSsq18
When comparing temperature records of Davies and Myrmidon loggers high
agreement was revealed.
IGOSS data between the two relevant squares was compared in two ways (Table1 &
Table2). Correlation yielded a coefficient of 1, and subtraction the monthly SST of
the two squares for Jan, Feb, and Dec yielded only negligible differences. Therefore,
one square (17.5S-18.5S and 146.5E-147.5E) was chosen for the subsequent
comparisons and investigations.
Finally, comparison between IGOSS data and the local reef loggers also showed high
agreement.
3.4 ENSO events
During the time of the survey, 4 major ENSO events could be identified (Table 3
&Graph1).
ENSO
1982-83
1986-87
1991-92
1997-98
Table 3: ENSO
The four ENSO events did have very different affects on the SST (Graph 2-5) . The
82/83 event was associated with previous winter and spring cooling before late
summer and autumn anomalies went well into the positive. During the 86/86 El Nino
temperature also showed a previous winter cooling, but anomalies already rose above
zero in spring and stayed there till autumn. The 91/92 showed an extensive winter
cooling, and anomalies rose above zero in spring, similar to the 86/87 El Nino.
However, anomalies then dropped below zero in early summer and did not show
major deviations from the mean for summer and autumn. Anomalies during the 97/98
event showed the common winter cooling, but after a slightly cool spring, anomalies
became highly positive in December and developed into an extensive summer and
autumn warming.
Monthly December anomalies (December temperatures were used to assess impact of
temperature on species – see below) for the El Nino event were heterogeneous. They
were strongly positive for the 97/98 event, slightly positive in the 86/87 event, neutral
91/92 and negative in 82/83..
3.5. Influence of cyclones on SST
The data record on cyclones revealed that 8 relevant cyclones hit the area during the
period of the survey (Table 4).
Name of cyclone Period of existence Date it hit survey area
Charlie 21-Feb – 01-Mar 1988 28-Feb
Ivor 16-26-Mar 1990 24-Mar
Joy 18-27-Dec 1990 26-Dec
Celeste 26-29-Jan 1996 26-Jan
Gillian 10-12-Feb 1997 12-Feb
Ita 23-23-Feb 1997 24-Feb
Justin 06-23-Mar 1997 23-Mar
Tessi 1-2-Apr 2000 2-Apr
Table 4: Cyclones in vicinity of survey area
Investigation of daily anomaly SST of the cyclone year revealed quite different
anomaly events at the time of the cyclones. For most cyclones – namely Ivor, Charlie,
Ita, Celeste, Tessi and Gilian – anomalies only displayed a 3-5 day drop in
temperature. Two cyclones, however, had a much bigger impact:
When Joy hit the area, anomalies had already dropped below zero 11 days earlier on
the 15th-Dec and dropped sharply again on the 19th-Dec (Graph 6). They stayed
highly negative till the 7th-Jan when temperatures rose briefly, just to immediately
drop again till the end of January. Since these highly negative anomalies stretched
over more than 1/3 of December it certainly influence monthly SST. This can be seen
from the monthly SST of Dec used in the species charts where Dec 1990 displayed
the lowest temperature throughout the survey period.
Cyclone Justin coincidenced with an even more pronounced drop in anomalies. They
dropped on the 8th-Mar and remained highly negative till the beginning of May, when
anomalies rose to a higher but still negative level up till the beginning of August when
the record stopped (Graph 7).
The mean SST anomalies of the 8 cyclones showed a distinct signal for a cyclone
event: anomalies became negative at day –14 and dropped sharply from day –2 to
day +2. Temperatures then remained below 0 till day +20 with a second, less
pronounced drop around day 18 (Graph 8).
When excluding Justin and calculating the anomalies for the remaining 7 cyclones,
the general signal remained, even though some differences showed: anomalies
dropped below 0 at day –11 and made a similar if not as deep drop as for the 8
cyclones from day –2 to day +2. Unlike for the 8 cyclones though, anomalies did not
remain below 0 for the whole 20 days after the cyclone incidence, but rose again
above 0 at day +11. A second, less pronounced drop can also be observed around day
18, when anomalies became negative again (Graph 9).
When temperature fields were examined, a spot of negative anomalies could be seen
on the Australian north-eastern coast. This spot was not apparent in February, became
pronounced in March and April (coinciding with the appearance of Justin) and faded
again in May (Graph 10).
Graph 10: SST anomalies: blue colors depict negative anomalies, red positive
Spatial and temporal patterns of recruitment
Overall pattern
Proportion of species with abundance
9% 2%
<=10
36%
<=100
28% <=1000
<=10000
<=100000
25%
Graph 11: Proportion of species in their abundance classes
family Spp Spp Spp Spp Spp Spp ubiquitous total
unique unique unique shared shared shared
at Pan at RIB at Myr inner/mid inner/outer outer/mid
Acanthuridae 1 1 2 - - 2 - 6
Apogonidae 2 2 1 3 - - 1 9
Blenniidae - 1 - - - 1 - 2
Caesionidae - - - - - 1 4 5
Chaetodontidae 1 1 1 3 - - - 6
Labridae 2 6 3 2 - 9 7 29
Lethrinidae - - - 1 - - - 1
Lutijanidae 1 - - - - - - 1
Microdesmidae - - - - - 1 - 1
Monacanthidae - - - - - 1 - 1
Ostraciidae - - - 1 - - - 1
Plesiopidae - 1 - - - 1 - 2
Pomacanthidae - - 1 - - 1 2
Pomacentridae 2 6 6 8 1 12 6 41
Scaridae - - - 3 - - 3 6
Serranidae 2 - 2 - - - - 4
Siganidae 1 - - - - - - 1
Tetradontidae - - 2 - - - - 2
Total 12 18 18 21 1 29 21 120
Table5: species across the shelf - species solely in each category
family Total Spp Total Spp Total Spp ubiquitous total
at Pan at RIB at Myr
Acanthuridae 1 3 4 - 6
Apogonidae 6 6 2 1 9
Blenniidae - 2 1 - 2
Caesionidae 4 5 5 4 5
Chaetodontidae 4 4 1 - 6
Labridae 11 24 19 7 29
Lethrinidae 1 1 - - 1
Lutijanidae 1 1 - - 1
Microdesmidae - 1 1 - 1
Monacanthidae - 1 1 - 1
Ostraciidae 1 1 - - 1
Plesiopidae - 2 1 - 2
Pomacanthidae - 1 2 - 2
Pomacentridae 17 31 25 6 41
Scaridae 6 6 3 3 6
Serranidae 2 - 2 - 4
Siganidae 1 - - - 1
Tetradontidae - - 2 - 2
Total 55 89 69 21 120
Table6: species occurring at the reefs across the shelf
# of Species Average
encountered unique at number
species reefs of
species
Inner shelf: 11.16
Pandorra 55 12
Mid-shelf:
John Brewer 100 - 29.24
Lodestone 91 - 26.02
Rib 89 18 32.66
Outer shelf: 69 18 15.86
Myrmidon
Table 7
Pan Rib Myr
unique 22% 20% 26%
shared 40% 57% 43%
ubiquitous 38% 23% 31%
Table 8
Individual reefs – Pandora
Recruitment patterns at the inner reef were very diverse across the whole
period(Graph 12), and diversity of new recruiting species was fluctuating strongly
throughout the survey period (Graph 13). During the investigation it became apparent
that diversity indeed did decline strongly over the past 18 years. Adding a line of best
fit to the diversity/year plot revealed a high R2-Value of 0.46 and species declining
from about 15 to 8 (see more in-depth results below).
When investigating the fluctuations in species diversity in connection with December
temperature of the previous year, a strong visual co-variation became apparent.
Species numbers did match temperature peaks and drops, and covaried especially well
between 1983 to 1994. After 1994, species numbers did match the overall peak in
temperature, but lagged the temperature rise in 1996 for one year and did show a drop
instead of only a declined in 1999.
Examination of data suggests that the two other environmental factors influenced
species diversity in 1996 and 1999. Cyclone Celeste hit the area in 1996 with heavy
rain. Wave action was high and this would seriously have affected the nearshore reef
Pandora which is at small depths, and could ultimately have affected the low
diversity. Furthermore, the reef might have experienced low salinity, but no data for
that is available.
Pandora seemed also to suffer from other cyclones. Recruitment pattern in 1991 is
very different from the other years (Graph 12). Cyclone Joy hit the survey area late in
1990 and affected a major flooding across the northern Queensland coast. Salinities
dropped as low as 28ppt nearshore, and the low salinity is likley to have acted on
species mainly recruiting close to the surface. Joy could therefore have shaped
recruitment pattern in 1991 to be so different from the other years. Other cyclones
including Justin did not have seemed to have e major impact on Pandora.
The environmental factor likley to have affacted diversity in 1999 was the major El
Nino event in 1998. This El Nino caused massive coral bleaching, with coral
mortality being 60% by 1999. Even though corals started to die extensivly at the
beginnig of March 1998, the recruitment survey at the end of March in 1998 did not
show a major change in species diversity or recruitment pattern (Graph 13).
However, by 1999 species numbers had dropped to an overall low. Investigation in
the recuit composition that year revealed that the Chaetodontids and Apogonids were
totaly missing, and all but two labrids were absent as well. While the three of the
surveyed chaetodontid species feed on Acropora, the coral that experienced the
highest mortality, and the fourth on hard corals in general, labrids are known to
mainly feed on crustaceans that need corals for shelter and food. Unfortunatley, only
two of the apogonids recorded at Pandora were identified to the species level, but
these, too, were crustacean feeders.
El Nino events in other years did not seem to have a similar influence on the species
diversity.
3.2.1 Midshelf
Midshelf reefs experienced much more similar recruitment patterns over the years
than the inner shelf (Graph 12). Like the inner reef, species diversity fluctuated in
accordance with temperature, but followed the actual temperature changes much more
closly. That was most apparent form 1984 till 1995. During the last years of the
survey where there was an overall temperature peak, divesity either followed this
trend weaky (at Rib Reef and JB) or with an overpronounced and a lagging peak
(LO).
On reason that species diversity did cease tofollow temperature closley might have
been cyclones. In 1996 cyclone Celeste went through the midshelf very close to Rib
and John Brewer reef. Similar to Pandora, the resulting waves could have killed
newly recruiting species and thereby preventing numbers from rising in accordance
with temperature. Since the northern slope of Lodestone reef was somewhat sheltered
from the cyclone by John Brewer Reef, species recruiting there would not have been
influenced by the cyclone.
Dropping species numbers at Lodestone and John Brewer, and only marginally rising
species numbers at Rib might also have been the result of influnced by cyclone Justin
in 1997. As pointed out above, the cyclone did lower SST significantley since the
beginning of March (Graph7) and recruits at the midshelf did experience these lower
SST for the 20 days prior to the species survey. Furthermore, Justin went through the
midshelf, and the resulting turbulences could have acted in a similar way as described
for Celeste.
Another cyclone with a major impact on the recruitment patterns was Joy. As in the
inner reef, recruitment pattern in 1991 did vary considerably from other years (Graph
12). Investigation into salinity levels due to the flooding revealed that salinity at the
midshelf dropped to abou 32ppt, which might readily have influenced the recruitment
pattern.
Alternativley, species diversity in 1998 and 1999 could have affected by the major El
Nino event in 1998. But since coral bleaching was not excessivley strong on the
midshelf, and coral mortality below 5%, this seems to be rather unlikley. There is no
indication for other El Nino event having had a major impact.
3.2.3 Outershelf
Similar to Pandora, the outer reef was diverse in their recruitment patterns across the
years (Graph 12). Species diversity did only flutuate to a limited degree, and this was
confined to 1983 to 1990. After 1990, species diversity did remain stable for about 5,
though composition of the recruits did not. Of the 41 species recorded through this
period, only 37% (15 species) were consistent (occured in 3 or more of the 5 years).
Those species came alomst exclusivly from the families Labridae (8 species) and
Pomacentridae (6 species). Opposedly, from the instable species, only 2 belonged to
Labridae, but 9 from the Pomacentridae.
Species level did rise again 1996, but since the 1997 recorsd was missing, statements
for those last years are rather vague. Species numbers did again drop in 1998-2000.
Cyclones did not have a major impact on Myrmidon. Unlike the inner and the
midshelf reefs, recruitment patterns in 1991 did not differ from the other recruitment
years. Due to the missing species record, the Justin in 1997 could not be assessed.
None of the other cyclones revealed a strong impact.
Neither did El Nino seemed have an influence on the reef. Coral bleaching in 1998
was low on the outer shelf reefs, and the resulting coral mortality close to not existent
(<1%).
1984 and 1989 were markedly different in their recruitment pattern (Graph 12) which
also is aparant in the species diversity. However, these results could not be readily
explained by any of the investigated environmental parameters.
Diversity decline at Pandora
Following the discovery of the diversity decline the recruitment patterns at Pandora
were investigated for species that ceased to recruit to the reef. Species condisidered to
fall into this category had to be recorded at least five times over the 18 year survey
and had to be markedly reduced in presence over the latter half. This investigation
revealed that 7 species stopped to recruit to the reef (albeit this number rose to 11
when species that occurred only three times, but were strongly clustered in the first
half of the survey period, were included) (Table 9). Species unique to Pandora were
equally affected as ubiquitous and shared species, and there was no obvious
connection to famliy. Only one of the species, Cheilodipterus quinquelineatus, did
also cease to recruiting to another reef across the shelf, which was Rib reef. All other
species where either unaffected in their recruitment across the shelf or could not be
evaluated since they were encountered only rarly (they were recorded only 1 or 2
times).
There was, however, a connection to feeding habits: except for one planktivorous
species and one species where diet could not be identified, all species were benthic or
benthic demersal crustacean or algae/weed feeders. Those two food types are readily
affected by community changes on the reef: a declining coral cover would affect a
wide range of crustaceans (Berkelmans, pers communication) in abundance, and some
factors could readily affect weed and algae cover.
Species Family Fate Diet where comment
Cheilodipterus Apogonidae Disappearing benthic Mainly in share Rib-
quinquelineatus crustaceans, corals
finfish
Labrid species Labridae Disappearing unique
Hemigymnus Labridae Disappearing benthic above ubi Rib+
fasciatus crustaceans, reefs Myr+
molluscs
Hemigymnus Labridae Disappearing demersal Juveniles unique
melapterus crustaceans among
branching
corals
Oxycheilinus Labridae Disappearing ubi Rib+
digrammus Myr sel
Acanthochromis Pomacentridae Disappearing phytoplants Coral ubi Rib+
polyacanthus & reefs, Myr sel
lagoons,
zooplankton
harbor
Neopomacentrus Pomacentridae Disappearing algae? above unique
bankieri /weed? reefs & at
outcrops
non-migr
Pomacentrus Pomacentridae Disappearing Benthic outer- share Rib sel
brachialis algae, slope
zooplankton Nonmigr
Scarus altipinnis Scaridae Disappearing Benthic Juv on share Rib sel
algae/weed protected
reefs
Diploprion Serranidae Disappearing benthic Near unique
bifasciatum crust & fish caves and
crev
Siganus Siganidae Disappearing sponges & Clear unique
punctatus tunicats lagoons
and reefs
–40m
Scarus sordidus Scaridae Appearing zoobenthos, Juv in Juv form
zooplankton, coral schools
detritus rubble that
areas migrate
hugh
distances
Table 9: Species changes at Pandora; species in italic were recroded only three
times across the survey period
# of Species Average % of
encountered unique number variation
species across the of
shelf species
Inner shelf: 66
Pandorra 55 12
Mid-shelf:
John Brewer 100 - 29.24 73
Lodestone 91 - 26.02 51
Rib 89 18 32.66 68
Outer shelf: 69 18 54
Myrmidon
ubiquitous 18
species with species Species that Proportion
response without could not be of species
response evaluated responding
per family
Acanthuridae 2 5 0
Apogonidae 2 2 6 50%
Blenniidae 1 1 0
Caesionidae 1 4 - ?
Chaetodontidae 1 2 3 33%
Labridae 15 10 12 60%
Lethrinidae 1 - 0
Luthianidae 1 - 100%
Microdesmidae 1 - 100%
Monacanthidae 1 - 0
Ostraciidae 1 - 0
Plesiopidae 2 - - 100%
Pomacentridae 15 15 17 50%
Scaridae 4 1 2 80%
Serranidae - - 4 -
Sigonidae - - 1 -
Tetradontidae - - 2 -
Table 5: Response of species grouped by family
Dec
John Brewer 0.27
Lodestone 0.20
Myrmidon 0.14
Pandorra 0.05
Rib 0.19
Table 6: Correlation coefficient for number of species with temperature at the
different reefs
ENSO 1982/83 - IGOSS sq17 anomalies & SOI
2.00
1.00
0.00
Jan-82 May-82 Sep-82 Jan-83 May-83 Sep-83
-1.00
Temperatur anomalies
SOI
-2.00
-3.00
-4.00
-5.00
Graph 2
ENSO 1986/87 - IGOSS sq17 anomalies & SOI
1.50
1.00
0.50
0.00
Jan-86 May-86 Sep-86 Jan-87 May-87 Sep-87
Temperature nomalies
-0.50
SOI
-1.00
-1.50
-2.00
-2.50
Graph 3
ENSO 1991/1992 - IGOSS sq17 & SOI
1.00
0.50
0.00
Jan-91 May-91 Sep-91 Jan-92 May-92 Sep-92
-0.50
-1.00
Temperature anomalies
-1.50
SOI
-2.00
-2.50
-3.00
-3.50
-4.00
Graph 4
ENSO 1997/98 - IGOSS sq17 anomalies & SOI
2.00
1.00
0.00
Jan-97 May-97 Sep-97 Jan-98 May-98 Sep-98
Temperature anomalies
-1.00
SOI
-2.00
-3.00
-4.00
Graph 5
Ivor, Joy 1990 - anomalies for MYR Logger 2
2.00
1.00
0.00 ano
1-Nov
1-Nov
1-Jan
1-Jan
1-Jul
1-Jul
1-Mar
1-Mar
1-Sep
1-Sep
1-May
1-May
-1.00
Ivor Joy
-2.00
Graph 6
Gillian, Ita, Justin 1997 - anomalies MYR Logger 2
2.00
1.00
0.00 anomalies
1-Jul
1-Jan
1-Jun
1-Feb
1-Oct
1-Mar
1-May
1-Nov
1-Dec
1-Sep
1-Aug
1-Apr
-1.00 Ita
Gillian Justin
-2.00
Graph 7
Mean impact of 8 cyclones on SST anomalies
0.20
0.10
0.00
0
4
8
12
16
20
-20
-16
-12
-8
-4
-0.10
-0.20
mean SST anomalies
-0.30
-0.40
-0.50
-0.60
-0.70
days before and after cyclone
Graph 8
Mean impact of 7 cyclones on SST anomalies
0.20
0.10
0.00
-20
-16
-12
0
4
8
12
16
20
-8
-4
-0.10
mean SST anomalies
-0.20
-0.30
-0.40
-0.50
days before and after cyclone
Graph 9
M93
M94
M91
ReP lot M86
M00
M87
M99
M95
++V M83
++H M98
1+2 M85
M90
P 99
PC P P 95
96 M96
SY M84
CV P 88 P 91 M92
P 94
P 93
P 00 P 98
P 83
P 85 M89
Y -V ars M88
P 97
P 87
Tran/S cale 86
PP 89
Dissim P 84
E NV P 92 P 90
RDA
P RDA
RDA -V ars R91
P RDA -V ars
Options R00 R88
Grps J91 R87
A dd-ons
L91 R90
R97
L00 R84 R92
R96
R95
J00 R85
R94
JB L84 L97 J97R86
R98
R89
J90
L90 R83
L87J87
LO
J84 J94 J98 J88R99
L86
L88 R93
MY L85
L99J89 L92
L98J86 J92
J95L89
L96
L83
L95
PA J83
J99 J96
RI J85
L94 L93 J93
Dim 2 24.93 %
Dim 1 37.65 %
Graph 12: ANOVA of the five reefs
PAN - average #spp & dec temp of prev year
2
18.00 29.50 R = 0.4551
16.00
14.00 29.00
12.00 #spp
28.50
10.00
dec temp of prev year
8.00
28.00 Linear (#spp)
6.00
4.00 27.50
2.00
0.00 27.00
1983
1985
1987
1989
1991
1993
1995
1997
1999
Graph 13
RIB - average #spp & dec temp
40.00 29.50
35.00
29.00
30.00
25.00 28.50 #spp
20.00
28.00 dec temp of prev year
15.00
10.00
27.50
5.00
0.00 27.00
1983
1985
1987
1989
1991
1993
1995
1997
1999
Graph 14
JB - average #spp & dec temperature
50.00 29.50
40.00 29.00
30.00 28.50 average #spp
20.00 28.00 dec temp of prev year
10.00 27.50
0.00 27.00
1983
1985
1987
1989
1991
1993
1995
1997
1999
LO - average #spp & dec temp
40.00 29.50
35.00
29.00
30.00
25.00 28.50 #spp
20.00
15.00 28.00 dec temp of prev year
10.00
27.50
5.00
0.00 27.00
1983
1985
1987
1989
1991
1993
1995
1997
1999
Graph 16
MYR - average #spp & dec temp
25.00 29.50
20.00 29.00
15.00 28.50 #spp
10.00 28.00 dec temp of prev year
5.00 27.50
0.00 27.00
1983
1985
1987
1989
1991
1993
1995
1997
1999
Graph 17
