close

Вход

Забыли?

вход по аккаунту

?

Fixed-station monitoring of a harbour wall community the utility of low-cost photomosaics and scuba on hard-substrata.

код для вставкиСкачать
AQUATIC CONSERVATION: MARINE AND FRESHWATER ECOSYSTEMS
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
Published online 6 October 2011 in Wiley Online Library
(wileyonlinelibrary.com). DOI: 10.1002/aqc.1230
Fixed-station monitoring of a harbour wall community: the utility
of low-cost photomosaics and scuba on hard-substrata
a
H. VAN REINa,*, C. J. BROWNa,b, D. S. SCHOEMANa, R. QUINNa and J. BREENc
Centre for Coastal and Marine Research, Environmental Sciences Research Institute, University of Ulster, Coleraine, Northern Ireland,
BT52 1SA
b
McGregor GeoScience Limited, 177 Bluewater Road, Bedford, Nova Scotia, Canada, B4BY 1H1
c
Conservation Science, Northern Ireland Environment Agency, Belfast, Northern Ireland, BT7 2JA
ABSTRACT
1. In marine temperate waters, the diverse epibenthic communities of hard substrata require robust, efficient,
high-resolution methods to better understand community dynamics and biological processes, such as growth,
recruitment and mortality that drive changes within communities.
2. The epibenthic community of a harbour wall was monitored from fixed stations over three seasons in 2009 to
evaluate the utility of a high-resolution photomosaic method. The community was described by visually estimating
cover and abundance of taxa within the photomosaics. In addition, areas occupied by the solitary cup coral
Caryophyllia smithii, the encrusting sponge Hymeniacidon perleve, and the red macroalga Plocamium
cartilagineum were digitized and their seasonal growth, recruitment and mortality estimated over the study period.
3. Significant changes occurred to the community in every season. The fewest number of taxa and the lowest
total taxa cover was observed in the winter. In the spring, there was significant growth of red macroalgae,
brown macroalgae, and ascidian taxa. In the summer there was significant growth of green macroalgae and turf
taxa, while red macroalgae experienced photoinhibition, and ascidians significantly declined in abundance.
4. The photomosaics enabled the extraction of data at a high taxonomic resolution, reduced the effects of
water-column turbidity, and through digitization, enabled estimation of the seasonal growth, recruitment, and
mortality of C. smithii, H. perleve and P. cartilagineum. However, the spring and summer canopies of macroalgae
and turf obscured the detection of many understory taxa, and thus, limited the accuracy of understorey
assessments. Provided the limitations of photomosaics are recognized, the method demonstrated in this study has
many potential benefits of advantage for fixed-station monitoring studies in temperate waters.
Copyright # 2011 John Wiley & Sons, Ltd.
Received 27 April 2011; Revised 24 August 2011; Accepted 28 August 2011
KEY WORDS:
sublittoral; monitoring; new techniques; photomosaics; benthos; macrophytes; invertebrates; growth; recruitment;
mortality
INTRODUCTION
The most dynamic changes in marine benthic communities
typically occur at fine scales (<1 m²), where biological processes
(i.e. growth, recruitment, mortality), ecological interaction
(i.e. predation, competition), and physiological tolerances
to environmental conditions have the greatest effect on the
status of those communities (van Rein et al., 2009).
Monitoring these parameters can reveal regime shifts
(Lundalv et al., 1986; Hewitt and Thrush, 2010), recovery
responses (Bell et al., 2006; Kroger et al., 2006), seasonal
changes (Alden et al., 1997), recruitment pulses (Goldberg
and Foster, 2002; Coles and Brown, 2007; Marzano et al.,
2007; Newton et al., 2007), growth pulses (Leukart, 1994)
and reproductive phases (Gaino et al., 2010) in their target
communities. Long-term monitoring at these scales can also
indicate the arrival of invasive species (Boets et al., 2011),
local pollution events (Hewitt and Thrush, 2010; Boets
et al., 2011) and climate change (Kroger et al., 2006).
Clearly, fine-scale monitoring of benthic communities yields
*Correspondence to: Henk van Rein, Centre for Coastal and Marine Research, Environmental Sciences Research Institute, University of Ulster,
Coleraine, Northern Ireland, BT52 1SA. E-mail: van_rein-h@email.ulster.ac.uk
Copyright # 2011 John Wiley & Sons, Ltd.
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
a plethora of ecological information of use to marine
managers worldwide (Diaz et al., 2004).
In Europe, there is an urgent need for monitoring at such
scales, where under the Marine Strategy Framework Directive
it has become a statutory requirement to collect high-quality
ecological information from marine substrata using
ecosystem-based approaches (EC, 2008; Cochrane et al., 2010;
Olenin et al., 2010). The Directive states that the current
status of the marine environment be known by 2012,
monitored regularly from 2014, and that protective measures
be implemented accordingly thereafter, so that Good
Environmental Status is attained in all marine habitats by
2020 (EC, 2008). To meet these requirements, monitoring
work has already begun (Borja et al., 2009; Ducrotoy, 2010;
van Hoey et al., 2010; Coggan and Diesing, 2011). However,
little of this effort has focused on subtidal communities of
hard substrata. Such is also an issue with work for the
Habitats Directive, under which regular surveillance of these
communities within Special Areas of Conservation has been
required since the 1990s (EEC, 1992). The aim of this study,
therefore, is to address this issue, and to build upon existing
efforts to develop robust, efficient fine-scale monitoring
methodologies for these Directives (Davies et al., 2001).
The techniques used in fine-scale studies are usually
developed by comparing methods to determine the optimal
balance between precision and accuracy, and cost and effort
(Brown et al., 2004; Jokiel et al., 2005; Leujak and Ormond,
2007; van Rein et al., 2011). Previous work indicates that
sampling from fixed stations is the most robust approach for
detecting changes in the benthos over time (Lundalv, 1985;
Kingsford and Battershill, 1998; Hill and Wilkinson, 2004).
On hard substrata, fixed stations are usually monitored by
scuba divers who collect data in situ from quadrats, either
through observation and hand mapping (Stone, 1970;
Carballo et al., 2008), or using optical-based methods, such as
video or stills cameras (Duckworth and Battershill, 2001;
Brown et al., 2004; Jokiel et al., 2005). However, the general
poor weather, cool temperatures, and low water-column
visibility at temperate latitudes have limited the use of optical
methods by scuba divers in temperate waters (Lundalv et al.,
1986; Bell et al., 2006).
The negative effects of water column turbidity (i.e. scatter)
can be limited by collecting imagery closer to the substratum,
although this reduces the overall sample area, additional
imagery can be collected from the surrounding area and
mosaicked to increase the final sample size (Burton et al.,
2007; van Rein et al., 2011). Indeed, image mosaics
encompassing 400 m² of sea bed have been used to assess
coral reef communities in the tropics (Lirman et al., 2007,
2010; Gleason et al., 2010). Furthermore, it is generally
agreed that optical methods collect data quickly, achieve
greater sampling objectivity than traditional diver
observations, generate permanent survey records and have
a low ecological impact in sensitive areas of conservation
importance (Bohnsack, 1979; Preskitt et al., 2004; Moore
et al., 2006; Barrett et al., 2007; Cabaitan et al., 2007;
Leujak and Ormond, 2007). As such, the mosaicking of
optical imagery represents a viable and suitable solution
to the issue of water-column visibility in temperate waters,
while possibly enabling the estimation of fine-scale
biological parameters of benthic communities, monitored
from fixed stations.
Copyright # 2011 John Wiley & Sons, Ltd.
691
In this study the utility of image mosaicking in a
fixed-station community-monitoring situation is evaluated.
The recommendations of van Rein et al. (2011) are closely
followed to do so: a study in which the effectiveness of a
low-budget, high-resolution image mosaicking technique
was assessed by comparing the relative efficiency of data
collection, extraction and analysis methods in a
multifactoral design. The final photomosaic method
recommended by van Rein et al. (2011) is trialled in a
harbour setting, as harbours are generally considered easy
marine systems to sample (Hewitt and Thrush, 2010). The
following two objectives were tested:
1. To quantify the established benthic community growing
on a quay wall and to describe community changes
over the seasons.
2. To estimate the biological parameters (i.e. growth,
recruitment and mortality) of solitary invertebrates,
colonial invertebrates and macroalgae from the
photomosaics at macrobenthic scales (> 1 mm) over the
seasons.
MATERIALS AND METHODS
Study site
The investigation was conducted in Portrush Harbour, located
on the north coast of Northern Ireland. The harbour is situated
on the western edge of Ramore Head (Figure. 1), with the ‘blue
flag’ waters of West Strand beach to the west and the rocky
reefs of The Skerries, with their rich epibenthic communities
to the east (Clements et al., 2010). This whole area is subject
to a tidal range of ~2 m, tidal flow of 1.5 m s-1, a high-energy
wave regime and seasonal fluctuations in water temperature
from ~7 C in February/March to ~14 C in August/
September (Atkins, 1997). It is also designated as a candidate
Special Area of Conservation (cSAC) because of its unique
assemblage of habitats and benthic taxa (Clements et al.,
2010). Portrush Harbour is situated within the cSAC and will,
therefore, require environmental protection through a
comprehensive monitoring and management regime. With a
maximum depth of 4.3 m and no commercial boat traffic, the
harbour provides an ideal site for regular monitoring
operations. Furthermore, as its breakwaters generally keep
turbidity low and hydrodynamic conditions relatively
stable, the harbour is also well suited to monitoring work
with optical methods.
The mature assemblage of epifauna and epiflora living on
the inner quay wall of the harbour was selected to test the
objectives of this study. Permanent stations were constructed
over five randomly chosen areas that could be monitored
seasonally. The permanent stations were constructed by
hammering marked climbing pitons into the wall at opposing
corners of 100 100 cm quadrat frames.
Community data collection and image mosaicking
The permanent stations were sampled during 2009, in February
(winter), May (spring) and September (summer). No autumn
sample was collected in December because the markers
from each permanent station could not be found in order to
resample the same areas along the harbour wall. The
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
692
H. VAN REIN ET AL.
Figure 1. Detailed illustration of the study site. (A) Location of Portrush within the United Kingdom (indicated by black dot). (B) Local bathymetry
and notable features surrounding Portrush Harbour (indicated by white box insert). (C) Close-up of the harbour indicating the inner quay wall where
the photomosaic samples were collected, with local bathymetry contours displayed in metres.
disappearance of the station markers was attributed to some
form of physical impact or abrasion along the entire site.
Therefore, no further monitoring of the community was
conducted lest we violate the fixed-station sampling design
of the experiment. In the seasons that were sampled
(henceforth referred to as winter, spring, and summer), the
samples were collected using the same photoquadrat
apparatus and method described by van Rein et al. (2011).
In summary, a Nikon 40 digital single-lens reflex (DSLR)
camera, in an Ikelite underwater housing with purpose-built
25 25 cm photoquadrat frame extending 40 cm outwards
from the lens, was used to collect sample images from within a
100 100 cm quadrat frame placed over each permanent
station. A 25 mm Nikkor lens was used to capture images on the
camera’s 10.2 megapixel charge-coupled device (CCD) sensor,
through an 8 inch Ikelite dome port. The 100 100 cm quadrat
frame was purposely divided into 16 equal areas, each
measuring 25 25 cm, to match the dimensions of the
photoquadrat frame. Once the 100 100 cm quadrat was held
securely in place between each station’s markers, sample images
(tiles) were collected from each of the 16 equal 25 25 cm areas
by a scuba diver. These image tiles, saved at a maximum
resolution of 300 dpi, were used to construct a single
photomosaic of each permanent station, at each season.
The photomosaics were constructed using the same low-cost
mosaicking method used by van Rein et al. (2011), with one
exception; an optical distortion filter was applied to the image
tiles to correct for any radial distortion in the imagery. The
method used Adobe Photoshop CS2 to crop each tile to show
only the 25 25 cm photoquadrat area. The cropped tiles
were then sharpened and the colour balance enhanced (where
necessary) before they were manually mosaicked into one
composite image of the entire 100 100 cm quadrat. Thus,
photomosaics 1 to 5 were created to represent each one of the
Copyright # 2011 John Wiley & Sons, Ltd.
permanent stations on the quay wall (Figure 2). During
the sample collection, it would have been impractical to
move aside portions of the canopy taxa in situ to better
sample the understorey biota (Dethier et al., 1993).
Therefore, in the interests of improving overall efficiency,
the communities observed in each photomosaic were
treated as two-dimensional when data were extracted
(Meese and Tomich, 1992).
Data extraction and analysis
Community structure
General community data were extracted from the photomosaics
using the visual cover estimation technique (henceforth referred
to as ESTIM) recommended by van Rein et al. (2011). This is a
well-established method used worldwide, where the observer
‘says what they see’ (Dethier et al., 1993; Leujak and Ormond,
2007; van Rein et al., 2011). In this study, all taxa within the
100 100 cm photomosaic area were identified to the highest
taxonomic level possible (i.e. species level) with the assistance
from the National Museums of Northern Ireland (NMNI), and
enumerated by either visually estimating their percentage cover
(for sessile modular/colonial invertebrates and macroalgal taxa),
or by recording their numerical abundance within the
photomosaic area. In some instances, taxa could be identified
only to coarser taxonomic levels and were thus recorded as one
of the following functional groups: red macroalgae (RF), brown
macroalgae (BF), green macroalgae (GF), crustose coralline
algae (CCA), sponges (SPON), hydroids (HYD), bryozoans
(BRY), ascidians (TUN) or mixed turf (T). These groups were
designed to reduce the overall community complexity while still
reflecting the general community structure of the quay wall
(Moore and Gilliland, 2000). As the observer (primary author)
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
693
Figure 2. Photomosaic sample imagery from quadrat 1 (100 100 cm), collected in winter, spring and summer 2009. White scale bar indicates 20 cm.
was sufficiently familiar with the species and functional groups
under investigation, learning effects during data extraction were
considered negligible.
To test the first objective, the general community data
were first divided into three groups to test the different
components of the community: all sessile colonial taxa
(recorded as percentage cover); all sessile and motile solitary
taxa (recorded as abundance); the functional groups (recorded
as percentage cover). The only factor considered in the
experimental design was SEASON. To minimize issues
associated with distributional assumptions of conventional
ANOVA, and to take advantage of the multivariate nature of
general community data, the PERMANOVA routine in
PRIMER-E was used to carry out the tests (Anderson, 2001;
Anderson et al., 2008). Before any testing, the expression of
dominant taxa was reduced and that of less dominant
taxa increased within the data by applying a square root
transformation to each batch of general community data
separately. Bray–Curtis similarity matrices were constructed
and a single-factor model was applied to each batch of
data, followed by post-hoc tests (PERMANOVA pairwise
comparisons), where appropriate. In each case, PERMANOVA
was run with 9999 unrestricted permutations. The components of
variation were estimated to provide a measure of multivariate
variability of the seasons within the groups of data. As a result of
using Bray–Curtis similarity matrices in their estimation, the
values in the components of variation can also be interpreted as
Copyright # 2011 John Wiley & Sons, Ltd.
percentage dissimilarity of conditions within experimental factors
(Anderson et al., 2008).
Follow-up PERMANOVA tests were conducted on the
following individual functional groups: red macroalgae, brown
macroalgae, green macroalgae, crustose coralline algae,
sponges, tunicates, and mixed turf. In addition, further tests
were conducted using the mean number of taxa per
photomosaic and the total space occupied by all sessile taxa
(i.e. colonial invertebrates and macroalgae) in each
photomosaic, in each season. None of these data were
transformed because only single functional groups were tested.
Biological processes
Measures of the biological processes (i.e. growth, recruitment,
mortality) of three types of taxa were determined by recording
the image area occupied by individual organisms over the
seasons. Two sessile invertebrates, the solitary cup coral
Caryophyllia smithii (Stokes and Broderip) and the encrusting
sponge Hymeniacidon perleve (Montagu); and the red
macroalga Plocamium cartilagineum (Linnaeus), were selected
to represent a solitary invertebrate, a colonial invertebrate,
and a canopy-forming macroalga, respectively. To save time,
each taxon was sampled from only one photomosaic:
photomosaic 2 for C. smithii, photomosaic 3 for H. perleve
and photomosaic 5 for P. cartilagineum. In each photomosaic,
areas occupied by individuals, colonies or macroalgal fronds
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
694
H. VAN REIN ET AL.
were first digitized and then quantified using the image analysis
software, Image J, developed by the United States National
Institutes of Health. In areas where sponge colonies were
growing in close proximity to one another, they were grouped
together and called clusters. Similarly, macroalgal fronds
growing in close proximity were grouped together and called
clumps. Organizing the sponges and macroalgae in this way
facilitated the estimation of change between the seasons.
Using the areas occupied by each C. smithii individual, H.
perleve colony and P. cartilagineum frond in each season, the
following biological parameters were estimated to test the
second objective: growth by measuring the mean basal areas
of cup corals, sponge clusters, and macroalgal clumps
between seasons; recruitment by tracking the number of new
cup corals, sponge clusters,and macroalgal clumps between
seasons; and mortality by recording the number of missing
cup corals, sponge clusters,and macroalgal clumps between
seasons. An additional measure of growth was obtained for
H. perleve and P. cartilagineum by describing the spatial
movements of sponge colonies into clusters, and macroalgal
fronds into clumps between the seasons.
RESULTS
Community structure
In total, 46 taxa were recorded from the quay wall in Portrush
over the study period, 33 of which were identified to species
level, with an additional three to genus level. The remaining
10 taxa were represented by their functional group.
Differences in the cover and abundance of all taxa in the
community are presented in Table 1. PERMANOVA tests
show that that all three components of the community
differed significantly across seasons (Table 2), and that
communities from no two seasons were alike (Table 3).
The sources of community dissimilarity were explored
through analysis of the functional group data (Table 4). These
data had the lowest multivariate variability (i.e. low values for
components of variation) and, therefore, the highest precision
of all three groups of general community data analysed. Over
the study period, the following trends are visible among these
data: the coverage of red macroalgae (RF) and ascidians
(TUN) peaks in the spring; the coverage of brown macroalgae
(BF), green macroalgae (GF) and mixed turf (T) peaks in the
summer; the coverage of crustose coralline algae (CCA) is
highest in winter and steadily decreases to its lowest
percentage in summer; and the coverage of sponges (SPON)
remains relatively constant throughout the study period
(Figure 3). Among the less dominant functional groups, the
coverage of bryozoans (BRY) peaks in the spring and there is
little seasonal change in cover among the hydroids (HYD)
(Table 1). The coverage of SPON shows no significant
changes between seasons (Table 4).
The total space occupied by all sessile taxa within the
photomosaics increased from winter to spring to summer
(Figure 4), although only between winter and spring was this
increase significant (t = 4.08, P = 0.007, df = 8). Similarly, the
mean number of taxa per photomosaic also increased with
seasons, but overall these changes were not significant (pseudo
F = 3.51, P = 0.059, df = 2). The lowest values for total space
occupied and number of taxa per photomosaic occur in the
Copyright # 2011 John Wiley & Sons, Ltd.
winter (February). In the spring (May), increases in total
space occupied and number of taxa coincide with significant
increases in the cover of RF, BF and TUN, while CCA
significantly decrease in cover (Figure 3; Table 4). There is
little change between the total space occupied and the number
of taxa per photomosaic from spring to summer. However,
many significant structural changes occur within the
community over this period: there is a decrease in RF and
TUN, and an increase in GF and T cover in the
photomosaics (Figure 3; Table 4).
Biological processes
The biological parameters of the cup coral C. smithii, sponge
H. perleve, and red macroalgae P. cartilagineum were
estimated for photomosaics 2, 3, and 5, respectively, at a
spatial resolution of 0.04 cm² (0.2 0.2 cm) (Table 5). In
photomosaic 2, the population of C. smithii displays
seasonal trends similar to those observed in all the
photomosaics: the highest abundance is in winter (46 m-²),
the lowest in the spring (18 m-²), with a slight recovery in
the summer (20 m-²). There are 14 corals that persist from
winter to spring, three of which continue to persist into the
summer. Seasonal observation of these corals shows that
growth is slow but perceptible, and only occurs in the spring.
The only coral growth observed in the summer occurs among
those corals that recruit in the spring. The growth rates from
these seasons are very similar: 0.04–0.05 cm² month-1. In general,
recruitment was low in comparison with mortality over the
seasons, with only four and three new cup corals appearing in
spring and summer, respectively, relative to 32 and 13
disappearing in the same seasons, respectively. Interestingly, 12
individuals missing from the spring photomosaic are visible
again in the summer, reducing the spring mortality to only 20
missing cup corals.
Observation of individual colonies of H. perleve in
photomosaic 3 demonstrates that they grow within clusters
of colonies (Figure 5). The number of colonies per cluster
varied throughout the study period, but with the highest
number of colonies per cluster in the spring (2.5 colonies
per cluster), and with the most new clusters in the same
period, this appears to be a period of high recruitment for
H. perleve (Table 5). The total number of sponge clusters
in photomosaic 3 decreases over the study period, from
33 m-² in winter to 13 m-² and 8 m-² in spring and summer,
respectively (Table 5). This indicates that the period of
maximum mortality is between winter and spring, when 25
sponge clusters disappear, relative to only eight between the
spring and summer. The disappearance of the clusters affects
the total area covered by H. perleve in photomosaic 3, which
drops from 88.7 cm² in the winter to 51.6 cm² in the spring.
However, among those clusters that persist and recruit to the
local area, signs of growth are evident: the mean basal area of
each cluster increases from an average of 2.7 cm² in winter to
4.0 cm² and 7.8 cm² in spring and summer, respectively.
However, observation of individual colonies, rather than
clusters, shows that colonies only grew in the summer (1.67 cm²
colony-1 month-1). In this season, there is a slight increase in
the total area covered in photomosaic 3, which rises to
62.4 cm². The cluster in Figure 5(A) illustrates the general
changes to the H. perleve population in photomosaic 3 over
the study period: a reduction in the number of colonies per
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
695
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
Table 1. Mean cover (%) and abundance (number m-2) of taxa identified from photomosiac samples collected in Portrush Harbour over three seasons
in 2009, with standard deviations included ( SD). Taxa measured in abundance indicated by (A) next to taxa name. Red macroalgae (RF); crustose
coralline algae (CCA); brown macroalgae (BF); green macroalgae (GF); sponges (SPON); hydroids (HYD); cup corals (CARSMI); polychaetes
(POLY); crustaceans (CRUST); gastropods (GAST); bryozoans (BRY); ascidians (TUN); fishes (FISH); mixed turf (T)
Winter
Taxa group
Taxon
Functional group
Rhodophyta
Delesseria sanguinea
Drachiella spectabilis
Callophyllis laciniata
Chondrus crispus
Halarachnion ligulatum
Phyllophora crispa
Plocamium cartilagineum
Lomentaria orcadensis
Rhodomenia pseudopalmata
Heterosiphonia plumosa
Rhodophyta indet. foliose
Corallinaceae
Dictyopteris membranacea
Dictyota dichotoma
Laminaria saccharina
Halopteris filicina
Carpomitra costata
Phaeophyta indet. foliose
Ulva lactuca
Chlorophyta indet. foliose
Haliclona urceolus
Leucosolenia spp.
Hymeniacidon perleve
Halichondria bowebankii
Porifera indet crust. 1
Porifera indet crust. 2
Porifera indet crust. 3
Hydrozoa indet.
Caryophyllia smithii (A)
Pomatoceros triqueter (A)
Cancer pagurus (A)
Liocarcinus depurator (A)
Palaemon spp. (A)
Lacuna crassior (A)
Lacuna vincta (A)
Trivia arctica (A)
Gibbula cineraria (A)
Calliostoma zizyphinum (A)
Alcyonidium diaphanum
Bryozoa indet crust.
Clavelina lepadiformis
Didemnum spp.
Ascidia indet. colonial
Botryllus schlosseri
Aplidium punctum
Gobiusculus flavescens (A)
Mixed turf taxa
RF
RF
RF
RF
RF
RF
RF
RF
RF
RF
RF
CCA
BF
BF
BF
BF
BF
BF
GF
GF
SPON
SPON
SPON
SPON
SPON
SPON
SPON
HYD
CARSMI
POLY
CRUST
CRUST
CRUST
GAST
GAST
GAST
GAST
GAST
BRY
BRY
TUN
TUN
TUN
TUN
TUN
FISH
T
Phaeophyta
Chlorophyta
Porifera
Cnidaria
Annelida
Crustacea
Mollusca
Bryozoa
Ascidia
Chordata
Unknown
Spring
Summer
Mean cover/abundance ( SD)
3.1 ( 1.6)
0.1 ( 0.1)
0.1 ( 0.1)
1.6 ( 2.6)
2.8 ( 1.8)
0.1 ( 0.1)
0.1 ( 0.1)
0.1 ( 0.1)
6.8 ( 2.0)
5.2 ( 2.8)
0.1 ( 0.1)
0.1 ( 0.1)
0.1 ( 0.1)
3.6 ( 2.3)
0.55 ( 1.0)
0.5 ( 0.6)
0.1 ( 0.1)
0.3 ( 0.3)
0.1 ( 0.1)
1.4 ( 0.4)
1.2 ( 1.0)
0.1 ( 0.1)
0.1 ( 0.1)
0.1 ( 0.1)
0.2 ( 0.2)
32.4 ( 14.3)
0.2 ( 0.4)
0.2 ( 0.4)
0.2 ( 0.4)
0.2 ( 0.4)
0.1 ( 0.1)
0.1 ( 0.1)
0.2 ( 0.4)
13.2 ( 1.8)
4.7 ( 3.0)
0.1 ( 0.1)
0.5 ( 0.7)
4.8 ( 3.7)
0.2 ( 0.3)
13.6 ( 3.5)
2.4 ( 2.0)
0.2 ( 0.4)
0.4 ( 0.3)
1.9 ( 4.2)
5.6 ( 1.3)
2.3 ( 2.3)
0.1 ( 0.1)
0.2 ( 0.2)
0.2 ( 0.2)
0.2 ( 0.1)
1.2 ( 0.6)
1.0 ( 0.7)
0.1 ( 0.1)
0.1 ( 0.1)
0.3 ( 0.3)
9.4 ( 5.1)
0.4 ( 0.5)
0.2 ( 0.4)
0.2 ( 0.4)
0.4 ( 0.4)
0.1 ( 0.1)
1.7 ( 0.5)
0.6 ( 0.5)
0.7 ( 0.9)
13.2 ( 2.6)
0.3 ( 0.2)
1.8 ( 2.1)
0.1 ( 0.1)
2.8 ( 2.8)
0.3 ( 0.4)
0.3 ( 0.5)
4.5 ( 1.7)
1.9 ( 1.4)
0.1 ( 0.1)
1.8 ( 1.3)
2.9 ( 3.6)
9.5 ( 1.1)
0.8 ( 0.8)
3.3 ( 3.1)
0.7 ( 0.3)
1.2 ( 0.6)
0.9 ( 0.8)
0.1 ( 0.1)
0.1 ( 0.1)
0.2 ( 0.2)
14.2 ( 6.2)
0.2 ( 0.4)
0.2 ( 0.4)
0.2 ( 0.4)
0.4 ( 0.5)
0.4 ( 0.5)
0.1 ( 0.1)
0.1 ( 0.1)
0.8 ( 0.4)
2.4 ( 0.9)
24.6 ( 2.3)
Table 2. PERMANOVA results for a single-factor PERMANOVA model, using the fixed factor SEASON, with 9999 unrestricted permutations. Tests
were conducted on root-transformed (a) percentage cover data from colonial invertebrate and macroalgal taxa, (b) abundance data from solitary
faunal taxa, and (c) percentage cover data of functional groups present in the 100 100 cm photomosaics. Bold results indicate a significant effect
(P ≤ 0.05). The components of variation (Comp. var.) indicate the multivariate variability of data between the seasons (SEASON), or between
replicates (Residual). perms = permutations
Data type
A. Colonial/algal taxa
B. Solitary taxa
C. Functional groups
Factor
df
SS
MS
Pseudo-F
P
perms
Comp. var.
SEASON
Residual
Total
SEASON
Residual
Total
SEASON
Residual
Total
2
12
14
2
12
14
2
12
14
4270.6
4568.2
8838.8
6160.5
5555.8
11716.0
1861.1
1052.7
2913.8
2135.3
380.7
5.6091
0.0001
9545
18.7
19.5
3080.2
463.0
6.6531
0.0001
9539
22.9
21.5
930.5
87.7
10.6070
0.0002
9583
13.0
9.4
Copyright # 2011 John Wiley & Sons, Ltd.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
696
H. VAN REIN ET AL.
Table 3. PERMANOVA pairwise comparison test results of the fixed factor SEASON. Tests percentage cover data from colonial invertebrate and
macroalgal taxa, abundance data from solitary faunal taxa, and percentage cover data of functional groups present in the 100 100 cm
photomosaics. Significance is indicated by bold type font, and its level by asterisks: *P ≤ 0.05, **P ≤ 0.01. WIN = winter; SPR = spring;
SUM = summer; perms = permutations
WIN/SPR
Data type
df
Colonial/ algal taxa
Solitary taxa
Functional groups
8
8
8
SPR/SUM
WIN/SUM
t-values
1.73**
2.33*
2.54**
perms
2.87**
2.72**
3.62**
2.46**
2.66**
3.57**
126
126
126
Table 4. (A) PERMANOVA results for a single-factor PERMANOVA model, using the fixed factor SEASON, with 9999 unrestricted permutations.
Tests were conducted on untransformed red macroalgae (RF), brown macroalgae (BF), green macroalgae (GF), crustose coralline algae (CCA),
sponges (SPON), ascidians (TUN) or mixed turf (T) functional group data, extracted from the 100 100 cm photomosaics. In each case, the
residual degrees of freedom (df) = 12. (B) PERMANOVA pairwise comparison test results of the fixed factor SEASON, conducted with the same
groups as in (a). Significance is indicated by bold type font, and its level by asterisks: *P ≤ 0.05, **P ≤ 0.01, *** P ≤ 0.001. WIN = winter;
SPR = spring; SUM = summer; perms = permutations
A. Main PERMANOVA
B. Pairwise PERMANOVA
WIN/SPR
Functional group
df
RF
BF
GF
CCA
SPON
TUN
T
2
2
2
2
2
2
2
Pseudo-F
4.60*
13.32***
7.16**
3.82*
0.23
6.92***
29.37***
perms
df
9284
9312
2208
5074
5636
5963
9612
8
8
8
8
8
8
8
SPR/SUM
WIN/SUM
t-values
2.18*
3.10**
0.86
2.22*
0.42
2.60**
0.13
2.75*
2.01
4.67**
0.45
0.37
6.87**
6.63**
perms
1.35
5.11**
2.72**
2.68*
0.68
1.52*
7.78**
126
126
66
66
66
56
126
Figure 4. Mean total space occupied (%) and mean number of taxa (S)
per 100 100 cm photomosaic in each season. Error bars show
standard error of the mean.
Figure 3. Mean percentage cover of functional groups in the 100 100
photomosaics at each season (as indicated). Error bars show standard
error of the mean. Red macroalgae (RF); brown macroalgae (BF);
green macroalgae (GF); crustose coralline algae (CCA); sponges
(SPON); ascidians (TUN); mixed turf (T).
cluster is counteracted by an increase in the size of those
colonies that persist.
Observation of individual fronds of P. cartilagineum in
photomosaic 5 shows that they form clumps as they grow,
consisting of many fronds together (Figure 5). The number of
fronds per clump varies only slightly over the study period;
however, when the total area occupied by P. cartilagineum
reaches its peak of 1081 cm² in the spring, the fronds within
each clump become so entangled that only one frond is
discernable per clump (Table 5). At this time, there are fewer
Copyright # 2011 John Wiley & Sons, Ltd.
clumps than at any other time in the study period, but each of
those clumps occupies a greater mean area than that of the
other two seasons put together: 120.1 cm² in spring vs.
39.5 cm² in winter and 49.0 cm² in summer. The spring is
evidently a time of massive growth for P. cartilagineum;
spring is the only season in which growth is observed among
those fronds that persist throughout the study period
(38.78 cm² frond-1 month-1). Despite the disappearance of four
clumps between winter and spring, two of these clumps
reappear in the summer, indicating that mortality in the
spring is more likely to be only two missing clumps.
Therefore, mortality remains low and does not change
throughout the study period. In the summer, however, there is
a decline in the total area of P. cartilagineum in photomosaic
5 to only 686.3 cm², despite this being the period of maximum
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
697
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
Table 5. Biological parameters of (A) the cup coral Caryophyliia smithii, (B) the sponge Hymeniacidon perleve, and (C) the rhodophyte Plocamium
cartilagineum measured from photomosaics 2, 3 and 5 respectively, over the seasons. Recruitment is indicated by new corals, new clusters and new
clumps; and mortality by missing corals, missing clusters and missing clumps of the cup coral, sponge and red macroalgae, respectively. Growth
rates are estimated for corals/clusters/clumps separately in three groups: those that persisted for the entire study period (WIN/SPR/SUM); those
that were present in only the winter and spring (WIN/SPR); those present in only the spring and summer (SPR/SUM). The number of corals/clusters/
clumps per growth group are indicated in italics. Standard deviations of all means are indicated within parentheses
Species
A. Caryophyllia smithii (Q2)
B. Hymeniacidon perleve (Q3)
C. Plocamium cartilagineum (Q5)
Biological parameters
Winter
Total corals (N)
Total area (cm²)
Mean coral area (cm².coral-1)
New corals (N)
Missing corals (N)
Reappearance of corals (N)
Mean growth rate (cm².coral-1.month-1):
WIN/SPR/SUM (3 corals)
WIN/SPR (11 corals)
SPR/SUM (2 corals)
Total area (cm²)
Number of sponge clusters (N)
Mean number of colonies (n.cluster-1)
Mean cluster area (cm².cluster-1)
New clusters (N)
Missing clusters (N)
Persisiting clusters (N)
Mean growth rate (cm².colony-1.month-1):
WIN/SPR/SUM (4 clusters)
WIN/SPR (4 clusters)
Total area (cm²)
Number of algal clumps (N)
Mean number of fronds (n.clump-1)
Mean clump area (cm².clump-1)
New clumps (N)
Missing clumps (N)
Persisiting clumps (N)
Reappearance of clumps (N)
Mean growth rate (cm².frond-1.month-1):
WIN/SPR/SUM (7 clumps)
WIN/SPR (1 clump)
46
16.94
0.37 ( 0.17)
-
18
8.68
0.48 ( 0.20)
4
32
-
20
10.01
0.50 ( 0.19)
3
13
12
-
0.05 ( 0.03)
0.04 ( 0.03)
51.6
13.0
2.5 ( 2.9)
4.0 ( 11.1)
5.0
25.0
8.0
No growth
0.05 ( 0.02)
62.4
8.0
1.4 ( 0.7)
7.8 ( 12.8)
4.0
8.0
4.0
473.8
12.0
1.4 ( 0.7)
39.5 ( 44.0)
-
No growth
No growth
1081.1
9.0
1.0 ( 0.0)
120.1 ( 107.4)
1.0
4.0
8.0
-
1.67 ( 2.33)
686.3
14.0
1.9 ( 2.0)
49.0 ( 69.1)
5.0
2.0
9.0
2.0
-
38.73 ( 28.68)
No growth
No growth
-
88.7
33.0
1.9 ( 2.8)
2.7 ( 7.6)
-
Spring
Summer
Figure 5. Digitized image areas representing the coverage of (A) one representative cluster of the sponge Hymeniacidon perleve and (B) one
representative clump of the red macroalga Plocamium cartilagineum measured from photomosaics 3 (Q3) and 5 (Q5), respectively, over the seasons.
Figure inserts show the positions of each cluster and clump within their parent photomosaic. The numbers of colonies/fronds within each sponge
cluster/algal clump are indicated with their mean areas beneath each image of seasonal cover.
Copyright # 2011 John Wiley & Sons, Ltd.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
698
H. VAN REIN ET AL.
recruitment (five new clumps). The photomosaic imagery shows
the relatively poor physical condition of algal fronds in this
period relative to the others, which further reflects this general
decline. The clump in Figure 5(B) illustrates the general
changes to the P. cartilagineum population in photomosaic 5
over the study period.
DISCUSSION
Using the photomosaic imagery collected from the quay wall in
winter, spring and summer, significant changes in the
community structure were evident and fully quantifiable.
Furthermore, accurate digitization of taxa was possible because
of high image resolution, which enabled estimation of the
growth, recruitment and mortality of the cup coral Caryophyllia
smithii, the encrusting sponge Hymeniacidon perleve, and the red
macroalga Plocamium cartilagineum within the photomosaics.
The relative ease of site access and data collection in every
season greatly assisted the fieldwork, while the natural seasonal
variability of the community highlighted the strengths and
weaknesses of the photomosaic method.
Benefits of using photomosaics
The key benefit of using the photomosaic method is that fine-scale
(< 1 cm) taxonomic features of small biota (~ 1 cm size) can be
identified and their spatial distributions mapped, quantified or
monitored over larger areas, such as 1 m² (van Rein et al., 2011).
The use of image mosaics in benthic ecological work is not a
new concept (Lirman et al., 2007, 2010), however, the use of
high-resolution stills imagery to construct the mosaics adds
further advantages to the accuracy and precision of data
extracted from the mosaics (van Rein et al., 2011).
Approximately 80% of the taxa detected in the photomosaics
were confidently identified to species level, including the sponge
H. perleve and the macroalga P. cartilagineum, known to have
fine-scale taxonomic features (Ackers et al., 1985; Bunker et al.,
2010). Those taxa not identified to species level in this study,
such as encrusting sponges and immature macroalgae, would
probably have been difficult to identify using even expert
observation in situ, let alone from imagery collected and viewed
in isolation (Bell et al., 2006). In these cases it is considered
acceptable to classify the taxa as functional groups (Drummond
and Connell, 2005; Blockey, 2007).
The inherent clarity of stills imagery is further improved
upon by collecting the imagery close to the subject (i.e. close
to the substratum), which in this study was from 40 cm
altitude above the substratum. It is accepted that capturing
images from altitudes of 30 or 40 cm can reveal cryptic taxa
better and aid the identification of species from the imagery
(Leujak and Ormond, 2007). Image resolution is typically lost
when communities are sampled from higher altitudes, such as
from ~ 2 m above the substratum, as this reduces the certainty
of taxonomic discrimination and the accuracy of cover
assessments (Leujak and Ormond, 2007; Lirman et al., 2007).
In our study, low-altitude sampling was a key component of
the mosaicking process and, therefore, any advantages it
conferred are considered a by-product of the specific mosaic
process. Although the issue was not investigated in this study,
water column turbidity is a fundamental problem for optical
methods in temperate waters (Barrett et al., 2007; Sayer and
Poonian, 2007). By virtue of the low-altitude nature of the
Copyright # 2011 John Wiley & Sons, Ltd.
sampling method in this study, however, the effects of water
column turbidity were reduced.
The digital imagery collected in this study also enabled easy,
accurate computer digitization of targeted taxa in the
photomosaics. Although considered an expensive method for
data extraction (Drummond and Connell, 2005), computer
digitizing comes recommended as a highly accurate method
for determining coverage of benthic taxa (Meese and Tomich,
1992; Whorff and Griffing, 1992; Pech et al., 2004). In this
study, it greatly facilitated the estimation of the growth,
recruitment and mortality of three targeted taxa that would
have otherwise been estimated from laborious drawing,
measuring and hand-mapping conducted in situ (Stone, 1970;
Duckworth and Battershill, 2001; Bell, 2002; Carballo et al.,
2008). As taxa were confidently digitized to areas measuring a
minimum of 0.2 0.2 cm, levels of accuracy rarely attained in
image-based benthic monitoring studies were achieved
(Duckworth and Battershill, 2001; Garrabou et al., 2002).
This enabled the complexities of H. perleve population
dynamics to be described to sufficient detail to highlight
similar trends to those described in a more exhaustive study
by Stone (1970). In this instance, sponges were laboriously
hand-mapped to illustrate growth dynamics, such as fusion
and fragmentation, and then coarse hand measurements were
made across only two axes of each sponge specimen (i.e.
maximum length and width) to determine colony areas and
subsequently growth (Stone, 1970). By comparison, greater
levels of accuracy were attained by digitizing the sponges in
this study without additional time spent sampling in situ.
Furthermore, the imagery was initially collected efficiently,
with low benthic community impact, and could be also stored
easily and indefinitely. These attributes are all considered
desirable features among the survey methods selected for use
in monitoring programmes (Brown et al., 2004; Jokiel et al.,
2005; Leujak and Ormond, 2007). Should further study of the
community become necessary a posterori, then the full
advantage of the permanent nature of optically collected
benthic samples can be realized, which is also of benefit to a
monitoring programme (Bohnsack, 1979).
Photomosaicking issues
Three main issues were identified with regard to use of the
photomosaics in this study. First, all imagery was subject to
lens-related optical distortion. Second, the movement of
canopy-forming taxa may have led to different representations of
community from moment to moment. Third, two-dimensional
(2D) representation of a three-dimensional (3D) community
may have led to the canopy-forming taxa obscuring those
taxa underneath. The first issue was successfully addressed
and corrected by applying a specialist filter to account for
any radial distortion in the imagery before mosaicking
(Choi et al., 2006). In addition, all imagery was checked
post-mosaicking to ensure that parallax errors along the
margins of each image tile were kept to a minimum and that
individual taxa were not represented twice. For the second
issue, however, there was no easy solution to a problem
which is commonly encountered in subtidal studies,
particularly those in which algal communities are moved by
tidal surge and current (Lundalv et al., 1986; Leonard and
Clark, 1993). This issue is considered a standard problem of
subtidal survey work with optical methods, one that must be
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
accepted as a general limitation of using optical imagery in
this type of community (Leonard and Clark, 1993).
The third issue relates the consequences of interpreting a 3D
community as though it were 2D. More specifically, cup corals,
crustose coralline algae (CCA), and some of the encrusting
sponges, such as H. perleve, showed decreased cover and
increased mortality in the spring and summer photomosaics,
at a time of expected growth (Leukart, 1994; Duckworth and
Battershill, 2001; Bell, 2002). Furthermore, there were 12 cup
corals that suddenly reappeared in the summer, following a
period of absence in the spring, in which they were assumed
to be dead. The most plausible explanation for these results is
that these understorey taxa were obscured from detection by
increased cover of macroalgal and turf taxa, growing in
response to increased periods of daylight (i.e. irradiance) and
water temperature typical of the spring and summer seasons.
A similar effect could be assumed for other understorey
taxa during these months. Therefore, as a consequence of
treating a 3D community as one that is only 2D, the
community in spring and summer was slightly misrepresented
in the photomosaics. This ‘canopy-obscure effect’ must be
considered as another limitation of using optical imagery for
monitoring 3D benthic communities. If not for the digitization
of the targeted taxa, which enabled accurate tracking of
individuals, this effect may have gone undetected resulting in
inaccurate estimations of seasonal growth and mortality for the
sponge H. perleve or the cup coral C. smithii.
It has become common practice when using optical
sampling methods in sublittoral benthic studies to treat 3D
benthic communities as though they were 2D (Preskitt et al.,
2004; Vroom et al., 2005, 2010; Page et al., 2006; Coles and
Brown, 2007; Walker et al., 2007). In temperate waters, where
communities typically have a macroalgal component, many
studies do not necessarily take measures to account for
potentially misrepresented understorey taxa, that may be
obscured from view by overgrowing canopies (Meese and
Tomich, 1992; Dethier et al., 1993; Pech et al., 2004). There
are a few potential solutions to this issue: sample only
mono-layered communities (Burton et al., 2007; van Rein
et al., 2011); sample each layer of the community separately
in multi-layered communities (Dethier et al., 1993); sample
using stereo-photography (Lundalv et al., 1986; Bell et al.,
2006); or sample the community at times of year when the
canopy will be less likely to obscure the understorey taxa.
This final suggestion is more applicable to studies aimed at
monitoring long-lived sessile taxa, such as those concerned with
many sponges and cup corals. The results suggest that the
winter, owing to low cover of macroalgae and turf, would be
the most suitable time of year to sample such taxa with optical
methods. However, owing to a lack of autumnal data or data
from any other seasonal samples, it is difficult to recommend
conclusively the optimal time of year to sample long-lived
sessile taxa because of potential temporal pseudo-replication
issues (Hurlbert, 1984).
Although it was not the purpose of this study to test
photomosaics in habitats other than those provided by
harbour walls, we can assume the canopy-obscure effect
would have been an issue consistent with such a study. As the
complexity of a 3D benthic community is affected by its
representation in 2D optical imagery, it is most likely that the
method would have collected imagery of poor quality from
structurally complex habitats of irregular bathymetry and
Copyright # 2011 John Wiley & Sons, Ltd.
699
high rugosity, such as those of boulder-slopes. Therefore, we
suggest that the photomosaic method should not be used as a
fixed-station monitoring method in habitats other than those
with a relatively flat surface, like harbour walls and vertical
drop-offs (Burton et al., 2007).
Seasonal changes in the quay wall community
Overall, the benefits of using the photomosaics to monitor the
quay wall community in Portrush harbour far outweighed the
constraints. The high-resolution imagery revealed general
changes in the structure of the community, as well as specific
changes to individual cup corals, sponge clusters, and
macroalgal clumps at macrobenthic scales (> 1 mm). These
changes reflect well-documented seasonal patterns of growth,
recruitment and mortality observed in other monitoring studies
of hard-substratum communities (Ojeda and Dearborn, 1989;
Leukart, 1994; Alden et al., 1997; Duckworth and Battershill,
2001; Marzano et al., 2007; Newton et al., 2007; Miller et al.,
2009; Vroom and Trimmers, 2009). However, the key difference
between these studies and this one is that here the imagery from
photomosaics was used to quantify these processes at a high
resolution, with relatively little effort and using relatively
standard sampling gear (i.e. photoquadrat). Furthermore, this
study highlights that time-consuming methods that require
microscope work, in situ hand measurements, in situ visual
observations, sampling with hand-operated sediment grabs and
dredges, or with closed chambers for productivity assessments
(Stone, 1970; Vethaak et al., 1992; Duckworth and Battershill,
2001; Bell, 2002; Miller et al., 2009; Gaino et al., 2010), are not
necessarily required to estimate growth, recruitment, and
mortality parameters of common benthic taxa. This adds further
to the potential of using low-cost photomosaics for fixed-station
community monitoring work (Brown et al., 2004; Hill and
Wilkinson, 2004, Bell et al., 2006).
The results showed the area occupied by the colonial sponge
H. perleve fluctuated little between the seasons, similar to the
mean cover of all the sponges throughout this study.
However, digitization of this species revealed a far more
complex pattern of recruitment, growth,and mortality over
the seasons, than that determined by the basic estimation of
cover. H. perleve populations are typically maintained more
by fragmentation than by reproduction because of their high
regenerative capacity (Stone, 1970; Gaino et al., 2010). Over a
typical year, the colonies that survive the winter become
reproductively active in the spring, timed to coincide with
increased water temperature and food availability; this phase
continues into the summer, when growth rates reach their
peak, and ends in late summer, after which a fall in water
temperatures coincides with the regression and fragmentation
of colonies through the winter, up until the following spring
(Stone, 1970; Bell and Barnes, 2002; Gaino et al., 2010). The
seasonal cycle of H. perleve observed in this study was similar
in many ways to this general seasonal cycle, with only a few
subtle differences: specimens occupied greater areas in the
winter, rather than the summer, and mortality was higher in
the spring, rather than the winter. These dissimilarities may
relate to the different water temperatures in Portrush, the
most northerly of all the H. perleve studies mentioned thus
far. However, it is more likely that many colonies were
hidden from view by the canopy during the spring and
summer, and thus, were present but not detected.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
700
H. VAN REIN ET AL.
Canopy-related obstruction is also the most likely explanation
for the apparent spring mortality event observed among the cup
corals C. smithii in Portrush Harbour. In similar habitats,
seasonal peaks in C. smithii mortality were observed in the
summer months (August and September) rather than in the
spring (Bell, 2002). However, it was suggested that old age and
the feeding activity of fishes were responsible for the observed
coral mortality in this instance (Bell, 2002). Without better
accounting for the canopy-obscure effect, it is difficult to
estimate reliably the biological parameters of C. smithii from the
photomosaics in this study. Despite these effects, however, the
digitization of those individual corals that persisted between
the seasons revealed growth and recruitment patterns consistent
with other studies (Bell, 2002; Moen and Svensen, 2004). These
individuals grew only in the spring, unless they had recruited to
the area in the summer, in which case they grew at the same rate
as those in the summer; at ~0.05 cm² month-1. The spring growth
probably coincided with the increased food availability typical of
this season (Alden et al., 1997), while the summer growth was
probably as a result of the rapid growth typically exhibited by
new recruits and younger corals < 4 mm in length (Fowler and
Laffoley, 1993; Bell, 2002). In this long-lived benthic species
growth is typically slow (Moen and Svensen, 2004) and
annual growth rates have been averaged to approximately
0.006 cm² month-1 year-1 (Bell, 2002). However, the seasonality of
growth detected in this study indicates that when food is more
available (i.e. in spring) the growth rate of C. smithii may
increase accordingly, similar to that observed in other
scleractinian species (Hamel et al., 2010). The digitization and
tracking of individuals has, therefore, proved a most useful tool
in estimating the growth dynamics of C. smithii.
Overall, the most reliable cover and biological parameter
estimates in this study were made of the macroalgae, a group
that was least affected by the canopy-obscure effect. All
macroalgal growth dynamics reflected general seasonal patterns,
even those described at macrobenthic scales, such as those of P.
cartilagineum. Because the photosynthetic pigments in red algae
(phycoerythrins) are highly sensitive to low light levels, these
algae can thrive in shade or at depth (Chapman, 1979). As such,
this particular group of macroalgae were the first to respond to
an increase in day length (i.e. increase in irradiance), and become
spatially dominant in the spring photomosaics. The massive
growth of P. cartilagineum in the spring clearly reflected this.
Individual fronds grew substantially over this period, becoming
completely entwined and only recognizable as single large
clumps. However, later in the year, between the spring and
summer, the phycoerythrins within the red algal fronds probably
reached saturation because of the increased irradiance, became
photoinhibited, and then limited any further growth (Chapman,
1979). The fragmentation of P. cartilagineum clumps into fronds
of poor physical condition echoed the effects of this
photoinhibition among the red macroalgae. The photosynthetic
pigments of the brown and green macroalgae, however, are
better adapted to function at higher irrandiances (Dring, 1981;
Vroom and Trimmers, 2009). Therefore, these macroalgae
overgrew the reds in the summer to become more dominant in
the photomosaics.
CONCLUSIONS AND RECOMMENDATIONS
There are clear advantages to using photomosaics for the
monitoring of hard-substratum benthic communities from
Copyright # 2011 John Wiley & Sons, Ltd.
fixed stations: fine-scale taxonomic features can be identified,
the basal areas of individual taxa can be digitized and their
biological parameters estimated to a spatial resolution of
0.04 cm² over areas larger than those possible with only single
photoquadrat images (Duckworth and Battershill, 2001).
Further advantages are gained in terms of sampling efficiency
and simplicity. This was revealed by the similarity between
the seasonal community dynamics estimated in this study and
those observed in other studies that used more laborious
manual sampling procedures in situ (Stone, 1970; Bell, 2002;
Garrabou et al., 2002; Carballo et al., 2008). Such advantages
are of particular benefit in Europe where robust, efficient
monitoring methods are in urgent demand to meet the
requirements of the Marine Strategy Framework and the
Habitats Directives (EC (2008) and EEC (1992) respectively).
The photomosaics have certain limitations, however, that
must be accounted for before their application in monitoring
work. Most notable is the ‘canopy-obscure effect’. This limits
the effectiveness of the method to habitats of relatively flat
profile, such as harbour walls and other vertical walls (Burton
et al., 2007). It also highlights the importance of selecting an
appropriate monitoring schedule for benthic communities in
relation to the seasons. The results show that reliable
estimations of macroalgal canopy cover can be made all year
round, while those of understorey taxa cover (i.e. cup corals,
encrusting sponges, and CCA) are less reliable as a result of
macroalgal overgrowth during the spring and summer
months. Consequently, we recommend that understorey
taxa be monitored only during periods of low canopy cover
(i.e. winter months) to reduce the inaccuracy of their
estimation. In addition, we stress that such monitoring
should be maintained over long time frames (>10 years) to
understand more fully the fine-scale community dynamics
of each site and how they change over time (Bell et al.,
2006; Coles and Brown, 2007). This understanding can lead
to better assessment of community condition at study sites
as well as providing a better indication of those changes
that are significant.
The utility of photomosaics could be further improved by
sampling over larger areas in these flat habitats, such as over
the same 2 2 m permanent plots used in other studies of
comparable nature (Stone, 1970; Duckworth and Battershill,
2001; Carballo et al., 2008). Fixed-station monitoring of such
areas could yield further insights into wider benthic
community dynamics, while maintaining the high image
quality and the possibility of species digitization demonstrated in
this study. Although loss of station markers is a problem
consistently experienced by fixed-station studies (Butler and
Connelly, 1999; Coles and Brown, 2007), many long-term
monitoring studies select sites with great care, use more robust
tools for marker installation (i.e. underwater drills and eye
bolts), and relocate those markers using sophisticated tracking
systems (i.e. ultra-short baseline tracking systems) (Whittington
et al., 2006). In combination, these efforts can result in highly
successful and rewarding monitoring of marine benthic
communities (Lundalv et al., 1986; Gaino et al., 2010).
ACKNOWLEDGEMENTS
The authors would like to thank Jan Coleman, Alex Callaway
and Rory McNeery for their valuable diving and fieldwork
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
support; Nigel McCauley for his technical support in the
development of the photoquadrat frame; Claire Goodwin
from the National Museums of Northern Ireland for assisting
in the taxonomic identification; and the harbour masters of
Portrush, Richard McKay and Angus Barry, for permission
to carry out the surveys. We are also grateful for the helpful
comments provided by the editor and reviewers which have
improved the flow and quality of the manuscript. The work
was funded by Northern Ireland Environment Agency
project entitled ‘Investigating monitoring methods for
assessing change in seabed habitats’ (University of Ulster
grant: 1203-R-0187).
REFERENCES
Ackers RG, Moss D, Picton BE, Stone SMK. 1985. Sponges of
the British Isles (“Sponge IV”). I. Marine Conservation
Society: Ross on Wye.
AFBI. 2011. Coastal Monitoring Network: (http://www.
afbini.gov.uk/index/services/services-specialist-advice/
coastal-science/coastal-monitoring.htm). [27/04/11].
Alden RW, Weisberg SB, Ranasinghe JA, Dauer DM. 1997.
Optimizing temporal sampling strategies for benthic
environmental monitoring programs. Marine Pollution
Bulletin 34: 913–922.
Anderson MJ. 2001. A new method for non-parametric
multivariate analysis of variance. Austral Ecology 26: 32–46.
Anderson MJ, Gorley RN, Clarke KR. 2008. PERMANOVA+
for PRIMER: Guide to Software and Statistical Methods.
PRIMER-E: Plymouth.
Atkins WS. 1997. Geology and physical environment. In
Coasts and Seas of the United Kingdom: Region 17,
Northern Ireland, Barnes JH, Robson CF, Kaznowska SS,
Doody JP, Davidson NC, Buck AL (eds). Joint Nature
Conservation Committee: Peterborough.
Barrett N, Edgar G, Lawler M, Halley V. 2007. A quantitative
video baseline survey of reef biota and survey of marine
habitats within Bathurst Channel, Southwest Tasmania
2002. Tasmania Aquaculture and Fisheries Institute report,
Tasmania.
Bell JJ. 2002. Morphological responses of a cup coral to
environmental gradients. Sarsia 87: 319–330.
Bell JJ, Barnes DKA. 2002. Density, distribution and decline
of two species of unattached demosponge. Sarsia 87:
110–118.
Bell JJ, Burton M, Bullimore B, Newman P, Lock K. 2006.
Morphological monitoring of subtidal sponges. Marine
Ecology Progress Series 311: 79–91.
Blockey D. 2007. Effect of wharves on intertidal assemblages on
seawalls in Sydney Harbour, Australia. Marine Environmental
Research 63: 409–427.
Boets P, Lock K, Goethals M. 2011. Using long-term monitoring
to investigate the changes in species composition in the harbour
of Ghent (Belgium). Hydobiologica 663: 155–166.
Bohnsack JL. 1979. Photographic quantitative sampling of
hard-bottom benthic communities. Bulletin of Marine
Science 29: 242–252.
Borja A, Bald J, Franco J, Larreta J, Muxika I, Revilla M,
Rodriguez JG, Solaun O, Uriarte A, Valencia V. 2009. Using
multiple ecosystem components, in assessing ecological status
in Spanish (Basque Country) Atlantic marine waters. Marine
Pollution Bulletin 59: 54–64.
Brown EK, Cox E, Jokiel PL, Rodgers KS, Smith WR, Tissot
BN, Coles SL, Hultquist J. 2004. Development of Benthic
Sampling Methods for the Coral Reef Assessment and
Copyright # 2011 John Wiley & Sons, Ltd.
701
Monitoring Program (CRAMP) in Hawai’i. Pacific Science
58: 145–158.
Bunker FSTPD, Maggs CA, Brodie JA, Bunker AR. 2010.
Seasearch Guide to Seaweeds of Britain and Ireland. Marine
Conservation Society: Ross-on-Wye.
Burton M, Lock K, Gibbs R, Newman P. 2007. Skomer
Marine Nature Reserve project status report 2006/07.
CCW Regional Report CCW/WW/07/4.
Butler AJ, Connelly RM. 1999. Assemblages of sessile marine
invertebrates: still changing after all these years? Marine
Ecology Progress Series 182: 109–118.
Cabaitan PC, Licuanan WY, Gomez ED. 2007. Comparison
between videographic and photographic methods in
assessing coral reef benthic communities. Science Diliman
19: 7–13.
Carballo JL, Vega C, Cruz-Barraza JA, Yanez B, Nava H,
Avila E, Wilson M. 2008. Short- and long-term patterns of
sponge diversity on a rocky tropical coast: evidence of
large-scale structuring factors. Marine Ecology - An
Evolutionary Perspective 29: 216–236.
Chapman ARO (ed.). 1979. Biology of Seaweeds: Levels of
Organisation. Edward Arnold: London.
Choi KS, Lam EY, Wong KKY. 2006. Source camera
identification using footprints from lens aberration. In
Digital Photography II, Sampat N, Dicarlo JM, Martin RA
(eds). SPIE Proceedings of Symposium on Electronic
Imaging: Science and Technology, Volume 6069.
Clements A, Plets R, Quinn R. 2010. Habitat Mapping of the
Skerries/ Causeway Proposed Marine SAC. Report
prepared for the Northern Ireland Environment Agency,
Belfast.
Cochrane SKJ, Connor DW, Nilsson P, Mitchell I, Reker J,
Franco J, Valavanis V, Moncheva S, Ekebom J, Nygaard
K, et al. 2010. Marine Strategy Framework Directive.
Guidance on the interpretation and application of
Descriptor 1: Biological diversity. Report by Task Group 1
on Biological Diversity for the European Commission’s
Joint Research Centre, Ispra, Italy.
Coggan R, Diesing M. 2011. The seabed habitats of the central
English Channel: A generation on from Holme and Cabioch,
how do their interpretations match-up to modern mapping
techniques? Continental Shelf Research 31: 132–150.
Coles SL, Brown EK. 2007. Twenty-five years of change in
coral coverage on a hurricane impacted reef in Hawai’i: the
importance of recruitment. Coral Reefs 26: 705–717.
Davies J, Baxter J, Bradley M, Connor D, Khan J, Murray E,
Sanderson W, Turnbull C, Vincent M. 2001. Marine
Monitoring Handbook. Joint Nature Conservation
Committee: Peterborough, UK.
Dethier MN, Graham ES, Cohen S, Tear LM. 1993. Visual
versus random point percent cover estimations: ‘Objective’ is
not always better. Marine Ecology Progress Series 96: 93–100.
Diaz RJ, Solan M, Valente RM. 2004. A review of approaches
for classifying benthic habitats and evaluating habitat
quality. Journal of Environmental Management 73: 165–181.
Dring MJ. 1981. Chromatic adaptation of photosynthesis in
benthic marine algae: An examination of its ecological
significance using a theoretical model. Limnology and
Oceanography 26: 271–284.
Drummond SP, Connell SD. 2005. Quantifying percentage
cover of subtidal organisms on rocky coasts: a comparison
of the costs and benefits of standard methods. Marine and
Freshwater Research 56: 865–876.
Duckworth AR, Battershill CN. 2001. Population dynamics
and chemical ecology of New Zealand Demospongiae
Latrunculia sp. no. and Polymastia croceus (Poecilosclerida:
Latrunculiidae: Ploymastiidae). New Zealand Journal of
Marine and Freshwater Research 35: 935–949.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
702
H. VAN REIN ET AL.
Ducrotoy JP. 2010. The use of biotopes in assessing the
environmental quality of tidal estuaries in Europe.
Estuarine, Coastal and Shelf Science 86: 317–321.
EC. 2008. Establishing a framework for community action in
the field of marine environmental policy. Directive 2008/
56/EC of the European Parliament and of the Council.
Official Journal of the European Communities 164: 19–40.
EEC. 1992. On the conservation of natural habitats and of wild
fauna and flora. Council Directive 92/43/EEC. Official
Journal 206: 7–50.
Fowler S, Laffoley D. 1993. Stability in Mediterranean–Atlantic
sessile epifaunal communities at the northern limits of their
range. Journal of Experimental Marine Biology and Ecology
172: 109–127.
Gaino E, Frine C, Guiseppe C. 2010. Reproduction of the
intertidal sponge Hymeniacidon perleve (Montagu) along a
bathymetric gradient. The Open Marine Biology Journal 4:
47–56.
Garrabou J, Ballesteros E, Zabala M. 2002. Structure and
dynamics of north-western Mediterranean rocky benthic
communities along a depth gradient. Estuarine, Coastal and
Shelf Science 55: 493–508.
Gleason ACR, Gracias N, Lirman D, Gintert BE, Smith TB,
Dick M, Reid RP. 2010. Landscape video mosaic from a
mesophotic coral reef. Coral Reefs 29: 253–253.
Goldberg NA, Foster MS. 2002. Settlement and post-settlement
processes limit the abundance of the geniculate coralline alga
Calliarthron on subtidal walls. Journal of Experimental Marine
Biology and Ecology 278: 31–45.
Hamel JF, Sun Z, Mercier A. 2010. Influence of size and
seasonal factors on the growth of the deep-sea coral
Flabellum alabastrum in mesocosm. Coral Reefs 29: 521–525.
Hewitt JE, Thrush SF. 2010. Empirical evidence of an
approaching alternate state produced by intrinsic
community dynamics, climatic variability and management
actions. Marine Ecology Progress Series 413: 267–276.
Hill J, Wilkinson C. 2004. Methods for Ecological Monitoring
of Coral Reefs. Australian Institute of Marine Science:
Townsville.
Hurlbert SH. 1984. Pseudoreplication and the design of
ecological field experiments. Ecological Monographs 54:
187–211.
Jokiel PL, Rodgers KS, Brown, EK, Kenyon JC, Aeby G,
Smith WR, Farell F. 2005. Comparison of Methods Used
to Estimate Coral Cover in the Hawaiian Islands. NOAA/
NOS NWHI Coral Reef Ecosystem Reserve Report,
Hawai’i.
Kingsford M, Battershill C (eds). 1998. Studying Temperate
Marine Environments. A Handbook for Ecologists.
Canterbury University Press: Christchurch.
Kroger K, Gardener JPA, Rowden AA, Wear RG. 2006.
Long-term effects of a toxic algal bloom on subtidal
soft-sediment macroinvertebrate communities in Wellington
Harbour, New Zealand. Estuarine, Coastal and Shelf Science
67: 589–604.
Leonard GH, Clark RP. 1993. Point quadrat versus video
transect estimates of the cover of benthic red algae. Marine
Ecology Progress Series 101: 203–208.
Leujak W, Ormond RFG. 2007. Comparative accuracy and
efficiency of six coral community survey methods. Journal
of Experimental Marine Biology and Ecology 351: 168–187.
Leukart P. 1994. Field and laboratory studies on depth
dependence, seasonality and light requirement of growth in
three species of crustose coralline algae (Corallinales,
Rhodophyta). Phycologia 33: 281–290.
Lirman D, Gracias NR, Gintert BE, Gleason ACR, Reid RP,
Negahdaripour S, Kramer P. 2007. Development and
application of a video-mosaic survey technology to document
Copyright # 2011 John Wiley & Sons, Ltd.
the status of coral reef communities. Environmental
Monitoring and Assessment 125: 59–73.
Lirman D, Gracias NR, Gintert BE, Gleason ACR, Deangelo
G, Dick M, Martinez E, Reid RP. 2010. Damage and
recovery assessment of vessel grounding injuries on coral
reef habitats by use of georeferenced landscape video
mosaics. Limnology and Oceanography: Methods 8: 88–97.
Lundalv T. 1985. Detection of long-term trends in rocky
sublittoral communities: Representativeness of fixed sites. In
The Ecology of Rocky Coasts, Moore PG, Seed R (eds).
Hodder and Stoughton Educational: UK; 329–345.
Lundalv T, Larsson CS, Axelsson L. 1986. Long-term trends
in algal-dominated rocky subtidal communities on the
Swedish west coast - a transitional system? Hydrobiologica
142: 81–95.
Marzano CN, Baldacconi R, Fianchini A, Gravina F, Corriero G.
2007. Settlement seasonality and temporal changes in hard
substrate macrozoobenthic communities of Lesina Lagoon
(Apulia, Southern Adriatic Sea). Chemistry and Ecology 23:
479–491.
Meese RJ, Tomich PA. 1992. Dots on the rocks: a comparison
of percent cover estimation methods. Journal of Experimental
Marine Biology and Ecology 165: 59–73.
Miller RJ, Reed DC, Brzezinski MA. 2009. Community
structure and productivity of subtidal turf and foliose algal
assemblages. Marine Ecology Progress Series 388: 1–11.
Moen FE, Svensen E (eds). 2004. Marine Fish and Invertebrates
of Northern Europe. KOM: Kristiansund, Norway.
Moore J, Gilliland P. 2000. Development of a monitoring
programme and methods in Plymouth Sound cSAC: Application
of diver techniques 1999. English Nature: Peterborough.
Moore CG, Saunders G, Mair JM, Lyndon AR. 2006. The
inauguration of site condition monitoring of marine features
of Loch Maddy Special Area of Conservation. Scottish
Natural Heritage Report: 152 (ROAME No. F02AA409).
Edinburgh.
Newton KL, Creese B, Raftos D. 2007. Spatial patterns of
ascidian assemblages on subtidal rocky reefs in Port
Stephens-Great Lakes Marine Park, New South Wales.
Marine and Freshwater Research 58: 843–855.
Ojeda FP, Dearborn JH. 1989. Community structure of
macroinvertebrates inhabiting the rocky subtidal zone in the
Gulf of Maine: seasonal and bathymetric distribution.
Marine Ecology Progress Series 57: 147–161.
Olenin S, Alemany F, Cardoso AC, Gollasch S, Goulletquer P,
Lehtiniemi M, McCollin T, Minchin D, Miossec L, Occhipinti
Ambrogi A, et al. 2010. Marine Strategy Framework Directive:
Task Group 2 Report: Non-indigenous species. Report for the
European Commission’s Joint Research Centre, Ispra, Italy.
Page HM, Dugan JE, Culver CS, Hoesterey JC. 2006. Exotic
invertebrate species on offshore oil platforms. Marine
Ecology Progress Series 325: 101–107.
Pech D, Condal AR, Bourgel E, Ardisson PL. 2004.
Abundance estimation of rocky shore invertebrates at small
spatial scale by high-resolution digital photography and
digital image analysis. Journal of Experimental Marine
Biology and Ecology 299: 185–199.
Preskitt LB, Vroom PS, Smith CM. 2004. A rapid ecological
assessment (REA) quantitative survey method for benthic
algae using photoquadrats with SCUBA. Pacific Science 58:
201–209.
Sayer MDJ, Poonian C. 2007. The influences of census
technique on estimating indices of macrofaunal population
density in the temperate rocky subtidal zone. Underwater
Technology 27: 119–139.
Stone AR. 1970. Growth and reproduction of Hymeniadicon
perleve (Montagu) (Porifera) in Langstone Harbour
Hampshire. Journal of Zoology 161: 443–459.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
FIXED STATION MONITORING USING SEASONAL PHOTOMOSAICS
Van Hoey G, Borja A, Birchenough S, Buhl-Mortensen L,
Degraer S, Fleischer D, Kerckhof F, Magni P, Muxika I,
Reiss H, et al. 2010. The use of benthic indicators in
Europe: From the Water Framework Directive to the
Marine Strategy Framework Directive. Marine Pollution
Bulletin 60: 2187–2196.
Van Rein H, Brown CJ, Quinn R, Breen J. 2009. A review of
sublittoral monitoring methods in temperate waters: a focus
on scale. Underwater Technology 28: 1–15.
Van Rein HB, Schoeman D, Brown CB, Quinn R, Breen J.
2011. Development of low-cost image mosaics of
hard-bottom sessile communities using SCUBA: comparisons
of optical media and proxy measures of community structure.
Journal of the Marine Biological Association of the United
Kingdom, DOI: 10.1017/S0025315411000233 .
Vethaak AD, Cronie RJA, van Soest RWM. 1992. Ecology
and distribution of two sympatric, closely-related sponge
species, Halichondria panacea (Pallas, 1766) and H. bowerbanki
Burton, 1930 (Porifera, Demospongiae), with remarks on their
speciation. Bijdragen tot de Dierkunde 52: 82–102.
Vroom PS, Trimmers MAV. 2009. Spatial and temporal
comparison of algal biodiversity and benthic cover at
Copyright # 2011 John Wiley & Sons, Ltd.
703
Gardner Pinnalces, Northwestern Hawai’ian Islands.
Journal of Phycology 45: 337–347.
Vroom PS, Page KN, Peyton KA, Kukea-Shultz JK. 2005.
Spatial heterogeneity of benthic community assemblages
with an emphasis on reef algae at French Frigate Shoals,
northwestern Hawaiian Islands. Coral Reefs 24: 574–581.
Vroom PS, Musburger CA, Cooper SW, Maragos JE,
Page-Albins N, Timmers MAV. 2010. Marine biological
community baselines in unimpacted tropical ecosystems:
spatial and temporal analysis of reefs at Howland and
Baker Islands. Biodiversity and Conservation 19: 797–812.
Walker SJ, Schlacher TA, Schlacher-Hoenlinger MA. 2007.
Spatial heterogeneity of epibenthos on artificial reefs:
fouling communities in the early stages of colonization on
an East Australian shipwreck. Marine Ecology 28: 1–11.
Whittington MW, Holt R, Irving R, Northen K, Stanwell-Smith
D. 2006. Across-Wales Diving Monitoring Project Volume 2:
Standard Operating Procedures. Countryside Council for
Wales Report 25b, Bangor.
Whorff JS, Griffing L. 1992. A video recording and analysis
system used to sample intertidal communities. Journal of
Experimental Marine Biology and Ecology 160: 1–12.
Aquatic Conserv: Mar. Freshw. Ecosyst. 21: 690–703 (2011)
Документ
Категория
Без категории
Просмотров
5
Размер файла
377 Кб
Теги
scuba, low, station, substrate, utility, cost, photomosaics, monitoring, community, fixed, wall, hard, harbour
1/--страниц
Пожаловаться на содержимое документа