Download Phytobenthic communities in the Baltic Sea succession

Phytobenthic communities in the Baltic Sea
- seasonal patterns in settlement and
succession
Susanne Qvarfordt
Department of Systems Ecology
Stockholm University
Stockholm 2006
Doctoral thesis 2006
Susanne Qvarfordt
Department of Systems Ecology
Stockholm University
S - 106 91 Stockholm
Sweden
Email: [email protected]
© Susanne Qvarfordt, Stockholm 2006
ISBN 91-7155-270-7 (pp. 1-39)
Cover photograph by Susanne Qvarfordt
Printed in Sweden by Universitetsservice US-AB, Stockholm 2006
To Isac
Abstract
Seasonal changes in reproduction, recruitment, occurrence and growth of
marine plant and animal species is a common phenomenon world-wide. This
thesis investigates whether such seasonal changes could determine the
succession in subtidal phytobenthic communities on free space in the low
diverse Baltic Sea. My results showed circular seasonal patterns both in the
settlement of species and in the annual appearance of communities. The
circular seasonal pattern was also observed in the succession. Initial species
assemblages were determined by the time space became available for
colonisation. Although the succession seemed to be directed towards one
site-specific final community structure determined by physical factors, the
time of the year when space became available influenced the rate of the
succession through species interactions. Rapid growth and timing of
settlement and free space occurrence allowed early species to occupy all
available space and prevent further colonisation, thereby slowing the
succession. My results also showed that both settlement and community
structure are influenced by substrate characteristics. Studying community
development on vertical artificial structures revealed communities with few
species and different composition compared to communities on vertical
natural substrates. A field study showed that settlement and community
structure changed significantly between 60º and 90º substrate slopes. This
thesis shows that some differences in the final community structure are
determined already at the settlement stage and that the succession pattern
varies depending on when free space occurs. However, small inter-annual
and site-specific differences in seasonal settlement periods and site-specific
final communities mainly determined by physical factors, suggest that
succession patterns are relatively predictable. Seasonal changes seem to
cause a spiralling succession towards a final, seasonally undulating, state.
List of papers
The thesis is based on the following papers, which are referred to in text by
their Roman numerals:
I.
Qvarfordt S., Kautsky H. & Malm T. (2006) Development of fouling
communities on vertical structures in the Baltic Sea. Estuarine,
Coastal and Shelf Science 67: 618-628.
II.
Qvarfordt S. & Kautsky H. Seasonal settlement of macroalgae and
sessile invertebrates in the northern Baltic proper. Manuscript
submitted to European Journal of Phycology
III. Qvarfordt S. & Kautsky H. Substrate slope and distance from the
seafloor, their effect on settlement and natural community
composition. Manuscript submitted to Journal of Experimental
Marine Biology and Ecology.
IV. Qvarfordt S. & Kautsky H. Succession following seasonal
establishment of hard substrate communities in the northern Baltic
proper. Manuscript.
Paper I is reprinted with the kind permission of the publisher.
My contribution to the papers included field preparation, sample collection
and sorting, data analysis and the main part of the planning and writing of all
papers. The studies were planned together with the co-authors, who were
also involved in the evaluation of the results.
Table of contents
Introduction
Community change
Colonisation
Succession in marine communities
Biotic interactions in the succession
Disturbance and free space
Seasonal variation
The Baltic Sea
Objectives of the thesis
Methods
Study areas
Field studies
Results and Discussion
Time scale in community development studies
Surface slope
Distance from the seafloor
Combined effects of several substrate characteristics
Comparison of artificial and natural substrates
Seasonality in settlement
Variable colonisation
Seasonal succession patterns
Seasonal effects on the rate of succession
Concluding discussion
Acknowledgements
References
Swedish summary/Svensk sammanfattning
1
1
2
3
3
4
5
5
7
9
9
10
12
12
14
15
16
16
17
20
21
22
24
29
30
37
Introduction
The shallow seafloor, the boundary between land and open sea, is as variable
in structure and climate as terrestrial land areas, creating a mosaic of
different habitats around our coasts that is home to a vast number of species.
In temperate rocky habitats, macroalgae and sessile invertebrates compose
the structure of the benthic communities, which is inhabited by mobile
organisms. The vegetation covered seafloor forms the phytobentic zone
which extends downwards until the light is too low to sustain net production
through photosynthesis. Highest on shore, the intertidal zone is the
intermediary between land and sea as it is periodically exposed to air due to
tidal changes in water level. The subtidal areas which are not subject to tidal
changes are constantly submerged. This thesis investigates the natural
development of subtidal phytobenthic communities after seasonal
disturbance in the Baltic Sea.
Community change
Succession is the sequence of changes in the biota that occur after
disturbance (Connell and Slayter 1977). Originally, succession referred to an
orderly and directional development of communities on space freed by
disturbance where each successive stage is dependent on the former, as
described by Clements (1916) and later modified by other authors (reviews
by e.g. Odum 1969, Drury and Nisbet 1973, Horn 1975). This so called
classical succession sequence includes early colonisers and later arriving
dominant species. The early species usually respond to r-selection having
characteristics such as rapid reproduction, wide dispersal and fast growth,
which make them efficient colonisers but generally poor competitors. The
later arriving K-selected species are characterized by slow growth and
limited dispersal but are good competitors. The arrival of a competitively
superior species eventually leads to decrease or extinction of the poorer
competitors on that space. The replacement of species continues until a
stable climax community is reached. The classical succession has been met
with criticism as many later studies have shown development where the later
stages are independent of earlier stages or even inhibited by early stages.
Studies in marine communities have shown little evidence of the classical
succession (Connell and Slatyer 1977), as numerous studies have shown
several, even contradicting, patterns of change (e.g. Fager 1971, Sutherland
1974, Osman 1977, Sutherland and Karlson 1977, Underwood and
Andersson 1994). However, colonisation sequences that indicate directional
change towards relatively stable final stages have also been observed (e.g.
Wilson 1925, Scheer 1945, Sousa 1979b, Dean and Hurd 1980, Mook 1980,
1
Turner 1983b, Farrell 1991, McCook and Chapman 1991). The variation in
the observed patterns shows that generalisations and predictions of the
development and final structure of any community should be supported by
studies in that particular community. I will use the term succession as
describing directional change in community composition over time,
observed during the community development on free space.
Colonisation
The first stage in the succession is the colonisation of free space.
Colonisation of free space generally occurs from three sources: via
vegetative growth from neighbouring areas, propagule dispersal or from the
spore bank (Ricklefs and Miller 1999). Most algae colonise new space by
generating propagules, reproductive spore bodies of different types that are
detached from the adult and have pelagic dispersion (Fletcher and Callow
1992). Algae can accomplish vegetative dispersal by developing shoots from
lateral outgrowths (Mathieson 1966). Vegetative reattachment of fragments
has been suggested to enable populations to persist in suboptimal
environments such as the northern Baltic Sea (Johansson and Eriksson
manus.). Algae may have more than one way of dispersal. In the Baltic Sea,
the red algae Ceramium tenuicorne (Kützing) Waern is capable of both
sexual and asexual reproduction as well as re-growth from vegetative
fragments (e.g. Bergström et al. 2003). In many marine habitats there is also
a bank of microscopic forms of seaweeds, as an adaptation for algae to
survive through stressful conditions (Santelices 1990).
Colonisation of free space in marine rocky habitats includes settlement of
sessile biota, which then form the biotic environment inhabited by mobile
fauna. Settlement is the point when an individual first takes up residence on
the substratum (Connell 1985). In marine benthic communities, recruitment
generally refers to recently settled juveniles that have survived for some time
after settlement (Keough and Downes 1982, Connell 1985), i.e. settlement
combined with early mortality. The length of this period of early mortality is
determined by the time of the first observation and thus recruitment is
defined by the observer, which means comparisons between studies may be
difficult. However, Connell (1985) concluded that it may be possible to use
densities of recruits as estimates of the density of settlers, which means that,
with the definition of settlement and recruitment in mind, comparisons
between studies can be made.
Studies have shown both temporal and spatial variation in propagule
composition in the water column (Hruby and Norton 1979, Amsler and
Searles 1980, Hoffmann and Ugarte 1985, Zechman and Mathieson 1985,
2
Santelices 1990) and in the settlement of planktonic propagules and larvae
(e.g. Haderlie 1969, Connell 1961, Hawkins and Hartnoll 1982, Keough
1983, Caffey 1985). This variation in propagule and larvae composition and
settlement may cause different colonisation sequences depending on when
and where space becomes available (Hruby and Norton 1979, Chalmer 1982)
as propagule availability determines the initial composition of species on
primary free space (Mook 1981, Caffey 1985). Recruitment that varies
spatially and/or temporally has been observed in intertidal and subtidal
communities world-wide (e.g. Dayton 1973, Foster 1975, Sutherland and
Karlson 1977, Sousa 1984, Moran and Grant 1989a, Menge 1991,
Underwood and Andersson 1994, Hutchinson and Williams 2001). Studies
have also shown that the success of invading species is dependent on the
resident species and thus early colonisers can influence community structure
long after their initial arrival (Osman 1977, Sutherland and Karlson 1977).
Succession in marine communities
There is an extensive literature on the succession in marine benthic
communities, especially from the intertidal shores which are more easily
accessible than the subtidal habitats. The studies have shown that community
succession varies greatly in space and time depending on a number of abiotic
and biotic factors, such as disturbance intensity, life history traits of the
species, density of herbivores and resident species (e.g. Sousa 1984,
Breitburg 1985, Turner et al. 1998, Benedetti-Cecchi and Cinelli 1993). Also
the time that space becomes available may influence the course of
succession due to temporal variation in species reproduction and growth
(e.g. Foster 1975, Emerson and Zedler 1978, Sousa 1979b, Hawkins 1981,
Turner 1983b, Dayton et al. 1984, Breitburg 1985, Serisawa et al. 1998).
Temporal variation in the settlement of species and in the occurrence of free
space would give rise to different initial species assemblages. Only those
species with settlement that coincides with the occurrence of free space,
would be able to colonise that space. Species that were not settling at that
particular time would be absent from at least a part of the succession. The
succession pattern would thus vary both between seasons and between years
due to seasonal and annual variation in the settlement of species (Chalmer
1982).
Biotic interactions in the succession
Variable colonisation of free space creates different assemblages of species,
whose interactions influence the succession. Three models have been
suggested to explain the sequence of species replacement on free space
(Connell and Slayter 1977). Facilitation is the mechanism behind the species
replacements in the classical succession theory. Early species alter the
3
abiotic or the biotic environment thereby facilitating the colonisation and
establishment of later arriving species, which are dependent on this
preparation. Although few marine studies have shown obligate dependence
on the presence of earlier species (Turner 1983b), several studies have
indicated that colonisation may be facilitated by earlier species (e.g. Menge
1976, Dean and Hurd 1980, Harms and Anger 1983, Lubchenco 1983).
Inhibition occurs when early species instead prevent colonisation of new
species and disrupt the succession. Species can inhibit colonisation by for
example pre-empting a limiting resource (Dayton 1971, Osman 1977) or
preventing spores from establishing contact with the substrate by barrier
effect, canopy characteristics or simply being swept away by whiplash
effects (Menge 1976, Ang 1985, Deysher and Norton 2003). Later arriving
species may also be tolerant of earlier colonists and their success
independent of the presence of already established species. In marine
communities, species are generally able to occupy all space and thus inhibit
new species from settling (e.g. Osman 1977, Sousa 1979b, Dean and Hurd
1980). However, mobile fauna can influence the interactions between sessile
species. For example, an inhibition effect on new recruitment caused by
early ephemeral algae occupying all available space can be broken by
selective grazing on the established early species. (Lubchenco and Menge
1978, Sousa 1979b, Lubchenco 1983, Kim 1997). Selective grazing can thus
facilitate colonisation of new species, but may also stall succession or
change the final community structure if the grazers instead feed on later
successional stages (see review in Sousa and Connell 1992).
Disturbance and free space
Free space is often a limiting resource in marine hard substrate communities,
as sessile species require an attachment surface for growth (Dayton 1971,
Osman 1977). Disturbances usually create secondary free space, i.e. space
which has remnants of its former inhabitants. The recolonisation can occur
directly from re-growth of remaining holdfasts (Lubchenco and Menge
1978, McCook and Chapman 1992) or via the spore bank containing micropropagules waiting for appropriate conditions (Santelices 1990). The
secondary succession is influenced by what survives the disturbance as these
species have a head-start on completely destroyed species (Kain 1975) and
the contents of the spore bank. Severe disturbances create space with less
remnants of former biota and the development will be more like primary
succession (Ricklefs and Miller 1999). Primary succession occurs on new,
completely empty, substrates, and the development of biotic communities is
dependent on colonisation from the outside. The substrates studied in this
thesis have been primary free space, as they were previously unoccupied by
biota. Succession in small patches of primary free space in otherwise
undisturbed areas is likely highly influenced by the proximity to neighbours.
4
Studies in many habitats have shown that the biota immediately surrounding
a clearing strongly influence the diversity and types of species that colonise
new space (Sousa 1984, Wahl 2001). For example, recolonisation of cleared
patches within existing subtidal communities in Australia occurred mainly
by vegetative growth from adjacent organisms (Keough 1984). The
neighbouring biota around a small patch is not only a source of colonisers
but may also diminish harsh sterile conditions on primary space.
Neighbouring biota can reduce water flow, provide shelter for grazers or
shadow the space, all important factors influencing the mortality of early
post-settlement stages (Vadas et al. 1992) and thus the succession on the
space.
Seasonal variation
Temporal and spatial variation in propagule composition and abundance
have been attributed to reproductive seasonality, fluctuating environmental
factors, the distribution of adults, the quantities of spores released and their
dispersability (Hruby and Norton 1979, Zechman and Mathieson 1985,
Hoffmann 1987). Marked seasonality in reproduction is a wide-spread
occurrence in temperate regions (Hoffmann 1987, Coma et al. 2000), and the
proportion of species showing seasonal variation in propagule release
increase with latitude (Santelices 1990). Laboratory studies have shown that
spore production in seaweeds is stimulated by light, temperature and nutrient
concentrations (Santelices 1990), factors which often fluctuate seasonally. In
regions with pronounced seasonal differences, such as the Baltic Sea,
seasonality in spore settlement may have a significant impact on community
development, as the colonisation sequence and the species interactions
would depend on the time free space occurs.
The Baltic Sea
Seasonal variations in the macroalgal communities in the Baltic Sea have
been described by Svedelius (1901), Waern (1952) and Wallentinus (1979).
Seasonal settlement patterns have been recorded for several phytobenthic
macroalgae in the northeastern Baltic proper (Kiirikki and Lehvo 1997) and
in the western Baltic Sea (Worm et al. 2001). No study has specifically
investigated seasonal effects on the succession in phytobenthic communities
in the Baltic Sea, although succession has been studied in microphytobenthic
communities (e.g. Kautsky et al. 1984) and in macroalgal and invertebrate
communities (e.g. Wahl 2001, Dürr and Wahl 2004, Enderlein and Wahl
2004). In the mid 1980’, the succession in rocky benthic communities was
studied at three depths in the Askö area (Fig. 1), northern Baltic proper
(Jansson B-O, Kautsky N and Wallentinus I. Successional developments in
rocky benthic communities in the northern Baltic proper, unpubl.). That
5
study showed that the succession in the Askö area was influenced by the life
history of the different species, interspecific competition and abiotic factors.
However, the final community structures were determined mainly by the
abiotic factors, such as depth, and reached after 2-5 years. The succession
was suggested to be more direct in the Baltic compared to marine habitats as
biological interactions are reduced due to fewer species. Also because many
grazers and predators, which may interrupt or redirect the succession are
absent. Even so, experimental studies have shown that grazing by small
invertebrates and interspecific competition between macroalgae, can
influence community structure and macroalgal diversity (Malm et al. 1999,
Lotze et al. 1999, 2000, Worm et al. 1999, 2001, Engqvist et al. 2000, 2004,
Berger et al. 2003, Raberg et al. 2005). However, the main determinants of
the structure of phytobentic communities in the Baltic Sea are still the
physical factors such as depth, substrate type and wave exposure as shown
by Kautsky and van der Maarel (1990).
It has been speculated that species-poor communities could have more
predictable succession sequences due to there being fewer paths of
community development available (Farrell 1991, Kautsky 1995). The unique
environment in the brackish Baltic Sea is characterized by the gradient of
decreasing salinity northwards and low species diversity together with
distinct seasons. Since it is naturally species poor, it allows investigations of
the succession on the community level in a relatively simple but fully
functional system. Studies on whole communities including all species
interactions make the results more comparable to reality than laboratory
studies or studies of selected parts of a community. In the case of succession
studies, which entail natural temporal variation and development on free
space, the results should be directly comparable to environmental monitoring
data. Monitoring programmes that regularly record the appearance and
composition of communities require knowledge of natural variation and
change in these communities to correctly identify unnatural change caused
by pollution or other anthropogenic disturbances (Christie 1985). A study of
succession in polluted and unpolluted fouling communities in Australia,
determined that pollutants removed sensitive species which played a key role
in the dynamics of the communities (Moran and Grant 1989b). Knowledge
of the pattern of community development may also allow for management
decisions that strive to minimize damage by planned disturbances and
facilitate recovery of former biota after disturbances. Timing submersion of
new structures or planned disturbances with the settlement periods of desired
species may shorten the recovery time.
6
Objective of the thesis
The aim of this thesis is to investigate seasonal effects on the recruitment of
species and on the succession. The main question addressed was whether the
development from different initial communities would be an orderly
succession to a specific final community, or if the development and final
community could be determined by timing of settlement and occurrence of
free space.
More specifically the following hypotheses were formulated:
•
Species-poor Baltic Sea communities develop towards a predetermined final structure after disturbance (paper IV).
•
There is seasonality in the settlement of species (paper II) and the
timing of available space and settlement periods creates differences
in colonisation and initial species composition on free space (paper
IV).
•
The final structure is determined by site-specific physical factors but
different initial species compositions could result in different
mechanisms governing the replacement of species in the succession
thus acting on the rate of succession (paper IV).
•
Stochastic effects from timing of settlement and occurrence of free
space can influence the community development (paper I and IV).
•
The community structure is determined by substrate characteristics
(paper I and III).
•
Differences in community composition between substrates can be
determined at the settlement stage (paper III).
An opportunity to study community development was presented when the
renovation of the pillars on the Öland Bridge was finished in 2001 (paper I).
The renovation produced homogenous substrates with communities of one to
eleven years of age, which were compared with regard to year and season of
submersion. In order to investigate the effect of seasonality on the
community development in a more controlled setting (paper IV), granite
plates were submerged on four occasions, once every season during one
year. Submerging the plates in different seasons was assumed to result in
different starting communities depending on which species were able to
settle at time of submersion. To determine which species were able to settle
at the time of submersion and to describe the settlement periods of species in
the area, the settlement was recorded during two years (paper II).
Communities on vertical artificial and natural substrate were compared
7
(paper I) and settlement on substrate with different slopes placed at different
heights over the seafloor were studied (paper III), in order to investigate
effects of some substrate characteristics.
8
Methods
Study areas
The two study areas, the Kalmar Sound and the Askö area are located along
the Swedish east coast in the brackish Baltic proper (Fig. 1.1). The tidal
amplitude in these areas is negligible but the water level is controlled by
weather conditions and deviations from the mean water level can last for
several weeks (Magaard 1974). Ice usually occurs during winter (Dietrich
and Schott 1974, Westring 1993, Westring 1995). The salinity is in the range
6.5 - 7.5 psu (Sjöberg and Larsson 1995, Tobiasson et al. 2001).
Fouling communities growing on the pillars of the bridge spanning the
central part of Kalmar Sound and on boulders in the vicinity were studied in
paper I (Fig. 1.2). The Kalmar Sound is 140 km long and 3 - 23 km wide,
located between the Swedish mainland and the Island of Öland (N56º, E16º).
The three other studies (paper II, III and IV) were carried out in the Askö
area (N58º, E17º) (Fig. 1.3) situated more to the north along the salinity
gradient present in the Baltic Sea. The four study sites were located in the
vicinity of the Askö Laboratory (Fig. 1.4). Seasonal settlement (paper II) and
succession (paper IV) were studied in two areas with flat rocky seafloor. Site
A was located along the shore of the larger island of Askö, and the more
wave-exposed site B at the northern tip of the smaller island Vrångskär. The
distance between site A and B was 1.5 km. The second settlement study
(paper III) was carried out in a flat rocky area (site S1) near site A. Natural
communities on rock surfaces with different slopes (paper III) was sampled
on site S2, which was located further northwest along the shore of the Askö
Island.
Plant and animal communities on hard substrates (paper II and III) including
the free-swimming mobile fauna (paper I and IV) have been investigated
using SCUBA-technique. The studies have been performed on hard substrate
between 1.5 - 4.5 m depth. At these depths Fucus vesiculosus L. often
dominates the plant biomass (Kautsky 1989). The flora and fauna associated
with F. vesiculosus include many ephemeral algae and mobile invertebrates.
At more exposed sites F. vesiculosus is mainly replaced by the red algae
Ceramium tenuicorne (Kütz.) Waern and Furcellaria lumbricalis (Huds.) J.
V. Lamour. (Kautsky 1989). Common sessile invertebrates are the blue
mussel Mytilus edulis L. and the barnacle Balanus improvisus Darwin.
Estimations for the Askö area state that M. edulis constitutes about 90 % of
the total animal biomass on hard substrates (Jansson and Kautsky 1977,
Kautsky 1989, Kautsky 1995).
9
The flat, granite rock at site A (paper II and IV) had 50 % coverage of Fucus
vesiculosus and 10 % coverage of Mytilus edulis. The remaining area was
mainly covered by filamentous algae; the most common were Pilayella
littoralis (L.) Kjellm. / Ectocarpus siliculosus (Dillwyn) Lyngb. (these two
brown algae are often difficult to separate and are then referred to as
Pilayella/Ectocarpus). The community at site B (paper II and IV) was
mainly composed of Furcellaria lumbricalis and M. edulis, which both had
50 % coverage. There were various filamentous algae of which Ceramium
tenuicorne was most prominent but only a few individuals of F. vesiculosus
(coverage <5 %). Site S1 (paper III) where effects of slope and distance to
the seafloor on settlement were investigated had 50 % coverage of F.
vesiculosus. The remaining area was mainly covered by filamentous algae,
of which Pilayella/Ectocarpus was most prominent. The animals M. edulis
and Balanus improvisus were also present on the rock. The rock slopes on
site S2 (paper III), where natural communities on different slopes were
sampled, were mainly covered by filamentous algae and M. edulis.
Field studies
Seasonal effects on succession were studied in paper I and IV. We took
advantage of a bridge renovation which provided substrates one to eleven
years of age submerged during all seasons. The fouling communities on the
bridge pillars were sampled once in July 2001, one year after the renovation
was finished. Communities of different age were compared with regard to
time of substrate submersion. In this study we had no control over the
experimental design, where and when the substrate was submerged which
produced some un-controlled factors. A controlled study of succession on
substrates available at different times over the year was therefore set up
(paper IV). Granite plates were submerged in July and October 2002 and in
January and April 2003 in the Askö area. Three plates from each plate-series
were sampled every third month for two years after submersion. In order to
observe development more closely, substrate submerged at the same time
was successively sampled in this study, contrary to the bridge study where
substrate was successively submerged and there only was one sampling. In
both studies whole communities, including both flora and fauna, were
quantitatively sampled using a quadrangular frame. The frame had one side
replaced with a net bag, mesh size < 0.5 mm, into which the contents were
scraped. This quantitative sampling method is used in the national
monitoring programme running in the Askö area and will effectively sample
sessile organisms and also mobile organisms as these tend to hide in the
sessile biota or take cover in the sampling bag. Species which are strongly
attached to the substrate or grow in crevices may be difficult to dislodge
fully but estimations of remaining biota were made in such cases. The
relatively smooth surfaces of the granite plates, the artificial concrete pillars
and the studied rock surfaces were easy to sample with this method. This
10
quantitative sampling method included species determination and dry weight
estimation of biomass for all species larger than 1 mm. Edge or border
effects on algal settlement have been observed as increased settlement within
1 cm of the edges of experimental blocks, likely due to water-flow patterns
(Foster 1975). In my study (paper IV), edge effects were expected to differ
between the plates as the water-flow patterns around the edges are probably
dependent on the immediate surroundings, i.e. shape of the rock surface and
identity of neighbours. In order to avoid edge effects 15 x 15 cm frames
were used to sample the 30 x 30 cm granite plates (paper IV). The size of the
quadrant frames used was otherwise 20 x 20 cm (paper I and IV).
The settlement periods of species were studied in order to determine which
species were able to colonise substrates available at different times over the
year (paper II). The settlement study was conducted parallel to the
succession study on the granite plates at the same two sites (A and B).
Settlement was studied from July 2002 to July 2004, i.e. during the first two
years of the succession study (paper IV), in order to be directly comparable
to the colonisation on the granite plates. A two year study length also
allowed some estimation of inter-annual variation in the seasonal pattern.
The round plastic discs (Ø 6 cm, area: 28 cm2), used as settlement substrates
had a rough surface intended to favour settlement. The discs were mounted
on bricks and exposed in two sets, with 1-month settlement discs exchanged
every fourth week and 2-month discs every eighth week in order to have
overlapping sampling periods. Germlings and larvae that settled just before
the sampling of the 1-month discs would have time to grow to visible size on
the 2-month discs. The discs were studied under a stereo-microscope and the
sessile species determined to species or closest taxa. Only a qualitative
estimate of abundance was made using a four-graded scale: scarce, present,
common or frequent.
The substrate in the succession studies (paper I and IV) differed both in
substrate slope and distance to the seafloor. In order to determine whether
colonisation is dependent on these substrate characteristics, settlement was
studied on different substrate slopes and at different heights above the
seafloor (paper III). The settlement discs were positioned at 0º (upper
horizontal surface), 30º, 45º, 60º, 90º, 135º and 180º on experimental stands
which elevated them eight or 40 cm above the seafloor. The height was
chosen in order to elevate the discs above the resident flora on the site but
still measure settlement at a height relevant to natural substrates. The
settlement at 40 cm distance from the seafloor could represent settlement on
top of small boulders. The discs were exposed in the field for four weeks.
Two experiments were run for a month each between May and July in 2003.
An area of 5.0 cm2 on each disc was quantitatively sampled by counting
11
algal attachment points and animal individuals or estimating coverage for the
bryozoans.
If settlement is dependent on substrate characteristics the final community
could be determined already at the settlement stage. In order to assess
whether settlement can determine final community structure, settlement on
different disc slopes were compared with natural communities on different
rock slopes. Natural communities were sampled on upper horizontal (0º)
rock surfaces and on 30º, 45º, 60º and 90º slopes at the end of June 2003.
The sampling was done using the 20 x 20 cm quadrangular frames (paper I
and IV) but biomass was only determined for sessile organisms and less
mobile animals such as mussels and snails, free-swimming animals were
excluded.
Studies have shown that type of substrate and surface texture can influence
recruitment and community structure (e.g. McGuiness and Underwood 1986,
Malm et al. 2003). In order to estimate how representative the observed
succession on the artificial bridge pillars were to natural vertical substrate,
the pillar communities were compared to communities on vertical sides of
five boulders in the vicinity of the bridge (paper I). The comparison between
pillars and boulders would also estimate how the introduction of new
substrate influenced the biodiversity in the area and show if the bridge pillars
provided habitats comparable to natural vertical substrate in the area.
Introduction of new substrate has been suggested to have positive effects on
the biodiversity as it increases the habitats available to sessile hard substrate
organisms.
Results and Discussion
Time-scale in community development studies
The time-scale is important to consider in studies of community
development, as the development of macroalgal and invertebrate
communities is a process that may take years. The final stage in the
development, when there is only small variation in the community structure
until a new disturbance occurs, is often difficult to determine as the different
stages in the succession can last for months to years. Connell and Sousa
(1983) proposed that community stability must be observed on time scales
consistent with the life histories of the resident species. Long-term
monitoring studies are naturally time-consuming and thus the bridge
renovation provided an excellent opportunity to study community
development (paper I). The renovation was conducted over a decade and
assuming that the age of the communities was the same as the pillar age,
community development could be compared over this time with just one
12
sampling (paper I). The time-scale in this study was consistent with the
recommendation of Connell and Sousa (1983), as the fouling communities
were composed of mainly annual species. Comparison of pillar communities
sampled in 2001 indicated that season of submersion influenced the
community structure during the first year as communities on pillars
submerged in different months during the year 2000 showed a comparatively
high dissimilarity (Fig. 2). The high similarity between older pillar
communities indicated that the communities on these artificial substrates had
reached a similar community composition after only one or two years.
Although, the species compositions on the older pillars likely vary within
and between years, due to the seasonal occurrence of the annual species,
these communities are the final communities on this vertical, artificial
substrate. Development of more stable communities including more
perennial species is not expected as perennial species that had not
established during these ten years seem unlikely to do so in the future.
Figure 2. Fouling communities on bridge pillars submerged in different years (paper
I). The separation from the other pillars and the larger variation among the pillars
submerged in 2000 is indicated by a dashed line. The years are shown and the
symbols shows the month of submersion. The multidimensional scaling was based
on Bray-Curtis similarity index and biomass, fourth-root transformed. The stress
value shows how well the two-dimensional plot describes the relation between
samples (stress values < 0.1 are good representations, values < 0.2 are useful but not
all details are correctly represented, values > 0.3 indicates that the data is not well
described by the plot).
Continuous monitoring of community development may explain more
variation than one sampling of different aged communities, as the history of
13
the community is known. In my study of succession following seasonal
establishment on granite plates (paper IV), the development was monitored
for two years after substrate submergence by continuous sampling. Timing
of settlement and occurrence of free space should be most important at the
beginning of community development. Thus, a time-scale of two years was
expected to show the impact of season on the succession and also allow
observation of the main patterns in community development, whether
communities converged towards one final community or not. Although the
study was not conducted over time-scales covering the full life-spans of the
species, following community development for two years showed clear
patterns in the succession due to season and site and there were indications
of both inhibited and facilitated colonisation. The settlement study (paper II)
was also performed over two years as it was meant to reflect the potential
settlement of organisms during the year and assess the inter-annual variation
in the settlement periods.
Surface slope (paper III)
The colonisation of free space is dependent on substrate slope (paper III).
The species composition changed with surface slope both on settlement discs
set at different angles and in natural communities on rock slopes, showing
that details of the final community structure could be determined already at
the settlement stage (paper III). Multivariate analysis including all settled
species showed different settlement on discs set at 0º- 60º compared to
overhang discs at 135º and 180º. The settlement on the vertical discs was
generally more similar to 135º and 180º discs (Fig. 3). Also, the natural
communities had different compositions on vertical rock surfaces compared
to the more horizontal surfaces, 0º- 60º (Fig. 3). The community structure
changed gradually on surfaces with slopes from 0º up to 60º. When the slope
increased further there were large changes in the structure of the epibenthic
communities, indicating a threshold between 60º and 90º. Algae dominated
on slopes up to 60º, whereas the vertical slopes had clear animal dominance.
Similar changes in community structure between slopes of 60º and 90º were
also observed on a breakwater in the Irish Sea (Chappel 1976 in Hartnoll
1983).
Furthermore, our results showed that some species may be excluded from
the developing communities already at the settling stage (paper III). For
instance, Fucus vesiculosus had low settlement success (< 1% of settled
germlings) on slopes steeper than 60º and are uncommon as adults on such
slopes. Thus, details of the final community structure can be determined by
mechanisms acting on the settlement. Observations of propagule
composition in the water column and settlement on shore (Hruby and Norton
14
1979) support this conclusion. Their study showed that only those species
normally found in the intertidal zone colonised the slides there, even though
other species were present in the water and thus should have been able to
settle.
a
Stress: 0.02
135
180 180
135
90
Experiment. 2
60
45
30
0
90
30
30
0
0
45
Stress: 0.1
0
0
60
45 30
90
b
0
90
30
45
90
90
45
60
60
40 cm
8 cm
30
60
60
Figure 3. a) Settlement on different substrates slopes and at different height above
the seafloor (paper III, Fig. 4). Settlement discs were set at 0º, 30º, 45º, 60º, 90º,
135º and 180º angles close to (8 cm) and elevated above (40 cm) the seafloor. The
settlement was recorded between 19.6 – 14.7.2003. The multidimensional scaling
was based on Bray-Curtis similarity index and abundance (square root transformed).
b) Natural community structure on different rock slopes, 0º, 30º, 45º, 60º and 90º,
sampled on the 26th of June 2003 (paper III, Fig. 2). The multidimensional scaling
(MDS) was based on Bray-Curtis similarity index and biomass, fourth root
transformed.
Distance from the seafloor (paper III)
Settlement also differed between substrate set close to and elevated from the
seafloor (paper III). Multidimensional scaling (MDS) based on all settled
species and including all discs shows the effects of both substrate slope and
distance from the seafloor (Fig. 3). The vertical separation in the MDS due
to distance to the seafloor was smaller than the horizontal separation at the
threshold, i.e. between 0º- 60º and 90º- 180º, thus showing the greater impact
of substrate slope. Only 40 cm elevation above the seafloor caused
significantly lower settlement of Fucus vesiculosus, Pilayella/Ectocarpus
and Cyanobacteria. These results are supported by Glasby (1999) who found
that the composition of subtidal epibiotic assemblages differed close to the
seafloor and 1.5 m above. This suggest that suspended substrates, often
employed in recruitment studies (e.g. Osman, 1977; Sutherland and Karlson,
1977; Dean and Hurd, 1980; Greene and Schoener, 1982) may not always
represent true benthic settlement in an area. Suspended substrate should for
example capture less of the large heavy propagules which in calm weather
tend to fall directly to the seafloor upon release from the adult plant, for
example Fucus vesiculosus (Serrao et al. 1997). The lower settlement on
15
substrates elevated 40 cm above the seafloor indicates lower recruitment on
top of boulders. However, effects of substrate elevation should probably be
considered together with water flow patterns around boulders, as this might
enhance settlement in a similar way as water flow around edges.
Combined effects of several substrate characteristics (paper I)
The communities on the renovated bridge pillars were impoverished even by
Baltic Sea standards, most likely as a result of the substrate slope, elevation
from the seafloor and artificial texture (paper I). The vertical bridge pillars
had high animal biomass, which corresponds with the slope of the substrate
(paper III). The pillar communities were also elevated from the seafloor,
which according to the results in paper III and Glasby (1999) can influence
the community structure. In addition, the bridge pillars were an artificial
substrate and the surface texture was expected to have an impact on
community composition. Artificial walls often lack microhabitats, such as
depressions, crevices and ridges, which are inhabited by many species
(Chapman 2003). The smooth vertical concrete structures, coupled with the
water movement and low salinity in the area, likely explains why few large
Mytilus edulis were observed on the pillars. Although, the biomass was
dominated by M. edulis, large mussels were found mainly on the horizontal
parts of the pillar foundations. Populations of Mytilus galloprovincialis
(Lamarck) were found to be unable to remain attached to concrete walls for
more than 2–4 years (Hosomi 1977). This effect of concrete vertical surfaces
is likely enhanced by the low salinity in the study area, as low salinity
hampers formation of byssus threads (Allen et al., 1976, Young 1985). This
would make M. edulis in the study area more easily detached by water
movement as they increase in size. Field observations have also observed
algae preferring rougher surfaces before smooth when settling (Harlin and
Lindbergh 1977). Although invertebrate larvae often are capable of some
active choice of their settlement substrate, the mobility of motile algal
propagules is negligible in comparison to the water motion (Norton 1985).
Algal preference for rougher surfaces is likely an effect of more depressions
and concavities that might catch the spores than smooth surfaces (Norton
and Fetter 1981).
Comparison of artificial and natural substrates (paper I)
The bridge pillars did not provide a habitat comparable to natural vertical
substrate in the area. Communities on the vertical sides of boulders included
several perennial algae whereas only one perennial alga, Polysiphonia
fucoides, was found on the pillars. The pillars also had higher biomass than
the boulders of P. fucoides and the most common animals, Balanus
improvisus and juvenile and small Mytilus edulis. These differences in
16
community structure indicate that the community development on the
artificial bridge pillars was not representative of natural vertical substrate.
This introduced new artificial hard substrate supported a community with
fewer species than otherwise found in the Kalmar Sound. These results agree
with Connell and Glasby (1999), who suggested that artificial structures are
not surrogate surfaces for communities occurring on nearby natural hard
substrates, even though they can increase biomass and biodiversity in some
cases.
Seasonality in settlement
The settlement varied with season as the sessile species in the area settle
periodically during the year (paper II). Highest settlement activity, both in
abundance and diversity, was observed in summer and early autumn (Fig. 4).
This is supported by studies in other cold-temperate regions, which have
shown that fertility for many species is restricted to summer and autumn
(Santelices 1990) when the major reproductive season generally occurs
(Hoffmann 1987).
No. of settled taxa
25
No. of settled taxa
Settlement activity
20
15
10
December
November
October
September
August
July
June
May
April
March
February
0
January
5
Figure 4. Summative settlement over the year in the Askö area (based on Table 1 in
paper II). The solid line shows the number of settled taxa. The dashed line show the
settlement activity, i.e. a cumulative abundance of settled propagules or larvae based
on the estimated abundance (scarce, present, common or frequent), which was
transformed into a value between 1 and 4. Low settlement activity is indicated when
the lines are close together whereas high settlement activity is indicated when the
lines are separated.
Multivariate ordination (MDS) based on the settlement of all species showed
that there was a circular seasonal pattern in the settlement. The circular
17
pattern was repeated in the second year of the study emphasising the
seasonality in settlement in this region. The design of the study did not give
the exact date when settlement started or ended, which meant the resolution
was limited to three or four weeks. However, the repeated pattern the second
year showed that, during the study time, the inter-annual variation in
settlement was shorter than four weeks. Thus, at a few metres depth in the
northern Baltic proper, there is a yearly cyclical variation in the settlement
and the small differences in the length and timing of settlement periods
between two study years indicated that the settlement may be predictable
within a few weeks. However, there is generally variability in propagule
aggregations both seasonally and inter-annually (Santelices 1990). For
instance, a study following fouling community development for 2.5 to 3.5
years in North Carolina showed seasonal patterns in larval recruitment and
dramatic differences in recruitment from year to year (Sutherland and
Karlson 1977).
The settlement patterns were also similar at the two study sites, even though
the distance between the sites was 1.5 km (paper II). Multivariate ordination
showed that the settlement was more similar between sites than between
seasons (Fig. 5). This showed that post-settlement processes mainly
determine the final community structure, as the more wave-exposed site B
had well developed Furcellaria-Mytilus-communities whereas the more
sheltered site A was situated in a Fucus-community. The high similarity in
settlement between sites indicated an evenly distributed spore and larval
supply in the area. Other studies have shown spatially and temporally
variable recruitment (Hawkins and Hartnoll 1982, Sousa 1984, Jernakoff
1985) even between sites located within 100s of metres (Benedetti-Cecchi
and Cinelli, 1993) or even 10s of metres (Hutchinson and Williams, 2001) of
each other. Location and fertility of the propagule source, i.e. the local flora
have been suggested to explain spatial patchiness in recruitment (Santelices
1990). For instance, local factors seemed to determine recruitment of
barnacles at different sites (Hawkins and Hartnoll 1982) and mussels may
create spatial patchiness through their filter feeding of spore-sized particles
(Santelices 1990). The immediate neighbouring biota, within 10 cm of
cleared patches, has also been shown to explain some variation in
recruitment (Sousa 1984). However, despite differences in the local flora and
fauna on the sites in my study, and the greater presence of mussels on site B,
the settlement at the sites was similar. My study shows that most species in
this area are able to colonise free space even though they are not common in
the immediate surroundings. That they do not compose a part of the
surrounding community shows that post-settlement factors mainly determine
their distribution in the area.
18
Comparison of settlement in my study included occurrence and a qualitative
scale of abundance. This may have enhanced the similarity both between
years and sites, as intra-specific differences in the amounts of spores
released are one of the principal factors determining spatial patchiness
(Santelices 1990). For example, a study of larval and adult macrofaunal
colonisation in Australia showed that temporal and spatial variation was
greater for measures of abundance of different species than for the diversity
of species (Chapman 2002).
Figure 5. The settlement pattern during one year-cycle on two sites in the Askö area
(paper II, Fig. 5). The numbers indicate month (1= January, 2=February etc), the two
sites are A and B and the year is indicated by empty triangles for 2002 and filled
reversed triangles for 2003. The multidimensional scaling was based on Bray-Curtis
similarity index and an estimate of abundance.
Although the settlement was similar between the sites in the study area,
studies from different regions of the Baltic Sea indicate differences in
settlement periods (Kiirikki and Lehvo 1997, Worm et al. 2001). The studies
were performed in different years which likely induce variation. Comparing
settlement pattern in my study with temperature and light data showed
relatively high correlations with both, although the factors are not
independent of each other. Studies have shown that climatic fluctuations in
temperature between years may induce inter-annual variation in the onset
and duration of settlement (Hoffmann 1987). The Baltic Sea spans latitude
54ºN to 66ºN, resulting in differences in the light and temperature climate,
19
which means inter-annual variations in settlement probably increase when
comparing settlement over larger areas. The species composition is also
dependent on the decreasing salinity gradient northwards. The differences in
settlement periods between regions in the Baltic Sea indicate that the
reproduction and dispersal of species is influenced by the stress of low
salinity and decreasing temperature and light gradients. Studies have shown
that latitudinal differences in temperature may influence the timing of the
settlement periods in species with wide distribution (see Hoffmann 1987 for
review).
Variable colonisation
Based on the consistent seasonal pattern in settlement both between years
and sites in the study area (paper II) we would expect the colonisation of free
space to be relatively predictable. The seasonal effects on colonisation were
apparent in the succession study (paper IV) when granite plates were
submerged during different seasons. Plates submerged in October had the
highest biomass of Pilayella littoralis, which has its peak settlement period
in December to January (paper II). Plates submerged in April had the highest
biomass of Fucus vesiculosus, which recruited in May-June on the study
sites (paper II). This means that the time that space become available can
determine colonisation success. Short-term effects of variable colonisation
were observed in the fouling communities on the bridge pillars (paper I). In a
community composed of annual species, variable colonisation, depending on
when substrate becomes available, should determine the species composition
mainly during the first year. Generally, most annual species decline after the
main growth season, and space available at different times during the year
should then be reset to a similar appearance. The colonisation and
development would resume from the same starting point and the community
structure would be determined by the yearly successive occurrence of annual
species and their abundance. This could be observed in the bridge pillar
communities which were composed of mainly annual species. The one-yearold communities on pillars submerged in different months during the year
2000 were separated from the other pillar communities, which formed a
year-mixed group (Fig. 2).
Effects of variable colonisation can be long-lived in a community, even in
the fouling communities on the bridge pillars (paper I). The only perennial
alga Polysiphonia fucoides occurred in higher biomass on pillars submerged
in autumn whereas the spring pillars had higher biomass of Balanus
improvisus. The settlement period of the barnacle begins in July (paper II)
and by then the spring pillars would have provided a suitable substrate for
barnacle settlement, with sufficient free space and little competition from
algae. Thus, available free space and coinciding reproduction season may
20
have favoured recruitment of barnacles on the pillars submerged in spring,
which in turn created long-term effects in the resulting communities. The
granite plates submerged in October (paper IV) were dominated by the
annual alga Pilayella littoralis, which seemed to inhibit colonisation by other
species on these plates. The effects of the timing of free space and peak
settlement period were still visible after 30 months. This result is supported
by Osman (1977), who found that even single rare changes, chance events,
in history can have long term effects on community composition. Also, a
study of short-lived fouling organisms showed that changes in older
communities were caused by invasion of new larvae and that their invasion
success was dependent on the colonisation history of the assemblage, thus
historical components could exist for long periods (Sutherland and Karlson
1977).
Seasonal succession patterns
The seasonality in settlement in the study area (paper II) influenced the
colonisation. The timing of settlement and free space influenced the
colonisation rate, determined the initial community structure and induced
seasonal effects on the succession (paper IV). Seasonal differences both in
the settlement (paper II) and in the occurrence of species (paper IV,
Wallentinus 1979) over the year created a circular pattern in the succession.
Disregarding the effect of sampling season, there was directional change in
community structure towards the surrounding community at each site,
although there were differences between the seasonal plate-series and
between sites. Despite similar settlement on both sites (paper II) the
surrounding rock communities and the succession patterns were different
(paper IV), showing that post-settlement processes mainly determined the
final community structure. Clear directional succession patterns towards
surrounding rock communities were observed in four of the eight
successional series. However, over time all plate communities became more
similar to the site-specific rock communities, indicating that there is one
final community structure. The other successional patterns or lack of patterns
may be explained by seasonal effects, i.e. different colonisation causing
rapid development of relatively stable communities due to timing of
substrate submersion and settlement activity, or different species
interactions, which act on the rate of succession. Similarly, studies have
shown that substrate available at different times over the year may undergo
different patterns of succession depending on the seasonality of species
reproduction (e.g. Foster 1975, Emerson and Zedler 1978, Chalmer 1982,
Turner 1983a, Breitburg 1985, Serisawa 1990, Underwood and Andersson
1994, Foster et al. 2003).
21
Seasonal effects on the rate of succession
The different initial species compositions on the plates resulted in different
species interactions acting on the rate of succession (paper IV). Rapid
colonisation of the July-plates due to summer being the main settlement
(paper II) and growth season led to quick establishment of many species, as
shown by the high diversity on these plates already after three months (paper
IV). The succession on the July-plates was probably slower the next spring,
as the established species seemed to inhibit colonisation by new species. For
example, Fucus vesiculosus occurred in lower biomass on July-plates
compared to the other plate-series. Slower succession rate was also observed
on the October-plates, where the dominating Pilayella/Ectocarpus seemed to
inhibit colonisation by Furcellaria lumbricalis, F. vesiculosus and Punctaria
tenuissima (C. Agarth) Grev. The presence of Pilayella/Ectocarpus in higher
biomass at site A than site B probably caused a slower the succession on the
plates there. Inhibition of new colonisation by the established species is a
common occurrence in marine communities (e.g. Dean and Hurd 1980,
Chalmer 1982, Harms and Anger 1983, Greene et al. 1983, Turner 1983a,
Underwood and Anderson 1994). Even early ephemeral algae are often able
to inhibit colonisation by later arrivals during their life-time (Foster 1975,
Sousa 1979b). The annual Pilayella littoralis is a common epiphyte in spring
and early summer, but by July most of the biomass has usually detached.
However, although the biomass on the plates decreased after the peak in
April, when P. littoralis is established directly on rocky substrate it seems
capable of occupying the space until the next settlement period, thereby
creating long-term effects in the community.
On the more wave-exposed site B, Mytilus edulis seemed to facilitate
colonisation of the red alga Furcellaria lumbricalis thereby accelerating the
succession (paper IV). This alga established successfully on most plates at
site B but occurred only sporadically on plates at site A. Dispersal of F.
lumbricalis in northern Baltic Sea occur mostly by reattachment of
fragments (Johansson and Eriksson in manus.). The higher abundance of M.
edulis on site B than on site A may have facilitated establishment by
fragments of F. lumbricalis (paper IV). The byssus threads of M. edulis have
been hypothesised to entangle fragments of algae, thereby giving the algae
longer time to produce new holdfasts and reattach (Waern 1952, Johansson
2002). I observed that F. lumbricalis seemed to have difficulties colonising
bare substrate as it was rarely observed at either site on the settlement discs,
despite being common in the surrounding community (paper II). The
settlement discs were sterile when submerged and thus presented a bare
surface without epibionts. A similar pattern where the colonisation of a
dominant species in the final community is facilitated by other species was
described for an intertidal community (Turner 1983b). She observed that the
late successional species surf grass Phyllospadix scouleri Hook. was
22
dependent on earlier species for successful establishment, as the barbed
spores became entangled in the early algae.
Facilitated establishment of Furcellaria lumbricalis together with higher
wave-exposure probably increased the rate of succession at site B. The most
rapid, direct succession towards the structure of the surrounding rock
communities occurred at site B, on space available for colonisation in
January and October (paper IV). Facilitation have been suggested be most
common in harsh environments (Connell and Slayter 1977, Turner 1983b).
The species dominating in the resident communities indicated higher waveexposure at site B than A, as Fucus vesiculosus generally dominate the plant
biomass but is at more exposed sites replaced by the red algae Ceramium
tenuicorne and F. lumbricalis (Kautsky 1989). The plate communities on the
two sites also became more different with increasing age, which indicates
that the physical factors, such as wave-exposure, are strong determinants of
the community structure in this area. The higher wave-exposure may allow
fewer species to survive and less competition for resources could favour
their establishment and increase the rate of succession. However, all species
that managed to establish on the granite plates in my study remained present
after two years, showing that no replacement of species had occurred.
Furthermore, few species in the local natural communities were present at
only one site, indicating that the differences were due to biomass distribution
between species rather than different species compositions. This is most
likely an effect of the few species in the area; most species are able to
establish but physical factors and to some extent biological interactions
determine how successful they are.
The size of the available free space can also influence the succession rate.
The effect of the size of free space is likely to vary depending on the species
composition in the surrounding communities. Observations have shown
variable recruitment in intertidal communities depending on proximity to
nearest conspecific adult, even in relatively small patch sizes, 25x25 cm and
50x50 cm (Sousa 1984). Species with short dispersal ranges, such as Fucus
vesiculosus (Serrao 1997), should have lower colonisation rates on large
patches of free space. Thus, the recovery time for Fucus-communities, for
instance site A, is likely to depend on the area disturbed. Also species that
colonise new space by vegetative growth should require more time to
colonise large patches of free space. Similar to vegetative growth, Mytilus
edulis often colonise space by crawling (Littorin and Gilek 1999) and its
invasion rate should thus be dependent on the size of the space. Slower
invasion of M. edulis could in turn also reduce colonisation by Furcellaria
lumbricalis, which may slow the succession at more wave-exposed sites,
such as site B. However, the recovery of smaller gaps may not always be
faster than larger ones, as shown in a study of mid intertidal communities in
23
the Mediterranean Sea (Benedetti-Cecchi and Cinelli 1993). They compared
the early succession in cleared patches, 12x12 cm and 22x22 cm, and
showed that algae were more abundant in the larger gaps, i.e. faster recovery
in larger gaps. This was explained by mode of colonisation, as no vegetative
recruitment was observed in the cleared areas. They concluded that
recolonisation by vegetative growth was probably more important in small
patches due to the longer perimeter length to patch area, whereas propagule
recruitment was probably more important for large patches (BenedettiCecchi and Cinelli 1993). In my studies, colonisation of new space seemed
to occur mainly through settlement either by propagules or fragments (paper
II and IV). Species with small easily dispersed propagules should be less
dependent on patch size whereas species dispersing by reattachment of
fragments are probably dependent on distance to adults and thus patch size.
Furthermore, studies have shown that the effect of patch size on the
succession is also dependent on the density of grazers (Sousa 1984,
Benedetti-Cecchi and Cinelli 1993). Sousa (1984) observed that the
influence of patch size on the abundance of grazers determined recruitment
success, as small patches had higher densities of grazers than large patches.
Although the material has not been completely analysed on the species level,
no obvious effects of grazing were observed in succession on the granite
plates (paper IV). However, Theodoxus fluviatilis L., one of two larger
grazing gastropods occurring in substantial quantities in the Baltic proper
(Skoog 1978), was common on newly submerged bare granite plates. The
gastropod T. fluviatilis is known to graze on the microphytobenthos (Skoog
1978), which may have influenced the initial species assemblages, as grazing
of microalgae can influence seaweed recruitment (Santelices 1990). The
snail has also been observed to graze on Fucus-germlings and high densities
may have an impact on Fucus recruitment (Malm et al. 1999). Grazing often
acts on the rate of succession (see Sousa and Connell 1992 for review).
Concluding discussion
The aim of this thesis was to investigate the effects of seasonal variation on
the recruitment of species and on the succession. There was a cyclic seasonal
pattern in the settlement (paper II) and occurrence of hard substrate species
in the northern Baltic proper (paper IV). Thus, the time that space becomes
available created differences in the colonisation and species composition on
free space. Timing of free space and settlement could create both short- and
long-term effects in the community structure (paper I and IV). Although, the
community structure was mainly determined by the physical environment
including such factors as substrate slope and texture (paper I and III), some
differences in the final community composition may be determined already
24
at the settlement stage (paper III). The development of phytobenthic
communities on hard substrates in the northern Baltic proper was directed
towards one final community structure, which was mainly determined by
site-specific physical factors, for example wave-exposure (paper IV).
However, seasonal occurrence of free space combined with the seasonal
variation in settlement could create different species interactions, which
influenced the succession pattern (paper IV). Although the succession was
investigated only in small patches of free space, the results suggest that the
recovery rate after disturbance can vary depending on when a disturbance
occurs.
The small differences in settlement found between years and sites in the area
suggest that the colonisation sequence on free space at a given time during
the year may be predictable. Although species interactions could influence
rate and succession pattern, the physical factors seemed more important in
determining community structure. Thus, as indicated in paper IV, also the
succession and the final community seem relatively predictable. These
indications of a predictable succession may allow management decisions.
From a management perspective we might wish to quickly regain the former
communities in an area after induced planned disturbance, such as dredging
or introduction of new substrate. Predictable development of epibenthic
communities would make it possible to plan disturbances or perhaps aid in
the recolonisation of a disturbed area. Knowledge of the seasonal
successional patterns and timing disturbance or introduction of new substrate
with settlement period of the desired species could shorten the recovery
time. If a Fucus-community is desired (and possible with regard to abiotic
factors), then freeing space in April would facilitate quick establishment and
faster succession than freeing space in October when the colonisation would
probably be inhibited by filamentous algae. This is supported by a study of
intertidal communities in the Mediterranean Sea which showed greater rate
of recovery in plots (size 22 x 22 cm) cleared in September than March
(Benedetti-Cecchi and Cinelli 1993). Although the plots both in their study
and mine were small, this indicates how the recovery rate may be influenced
by the time of year a disturbance occurs. The recovery after disturbance is
dependent on other factors as well, such as magnitude of disturbance,
isolation, species dispersal ranges, local grazing pressure etc, and timing of
free space will of course interact with these factors.
Based on my studies and conclusions, an attempt was made to describe the
succession in phytobenthic communities in the northern Baltic proper. The
development of the species-poor communities on hard substrates was
characterized by seasonal variation. The seasonality was apparent as an
annually reoccurring cyclic pattern both in the settlement of sessile
organisms (paper II) and in the composition of communities during different
25
seasons (paper IV). The seasonal variation is no strictly random variation, as
it is a yearly pattern. The seasonality added a circular pattern to the
successional sequence. If the occurrence of free space over the year is
random, chance effects would be induced in the succession. Due to the few
species present in the Baltic proper there are less species interactions and the
community structure seems to be more determined by site-specific abiotic
factors. Thus, there is generally one final community structure based on the
prevailing physical factors. The highest variation in composition is found in
young communities, as not all species have had the opportunity to colonise
yet. The impact of season in the succession decreased with increasing
community age. This general pattern, including directional development
towards one final structure and a circular seasonal pattern that decreases over
time can be illustrated in a model (Fig. 6), and observed in the succession on
the granite plates (paper IV) (Fig. 7). However, some species compositions
may create long-term seasonal effects in the succession (paper IV).
SUCCESSION
Winter
SEASON
Autumn
Spring
Summer
Start
Intermediate
Final stage
Figure 6. Illustration of the effect of season on the succession pattern. The effect of
season is perpendicular to the succession and indicated by shaded regions. The effect
decreases over time as the succession progresses towards the final stage. The black
dots represent the final community sampled different seasons. The stars represent
different aged communities during succession sampled every season following
beginning of colonisation (start).
Although the final community structure seemed to be determined mainly by
site-specific physical factors, for example wave-exposure (paper I and IV),
26
the season of substrate submersion influenced both the rate of the succession
and the species interactions (paper IV). In an area with discontinuous
settlement the time when space becomes available will decide which species
colonise first (Chalmer 1982). In the Baltic proper, free space in winter will
initially be colonised by the few species capable of recruiting during the cold
season. Space available during summer, at the height of the settlement
activity, will quickly be colonised by most species in the area. The initial
rate of succession is therefore slower on space available during the colder
months when fewer species settle than in summer.
3
Winter
Stress: 0.01
Spring
Autumn
6
12
9
15
C
C
18
C
24
Summer
21
C
SITE A: JANUARY-2003
Au
12
tum
Spring
n
Stress: 0.06
3
24
15
9
18
Winter
C
21
C
C
C
Summer
6
SITE B: JANUARY-2003
Figure 7. The successional patterns on substrate submerged in January 2003 on the
two study sites A and B (paper IV, Fig. 4). The circular effect of season is indicated
by the dashed arrows between surrounding rock communities (C) sampled different
seasons and the line tracing the path of succession. The age of the sampled
communities is given in months. The symbols show the sampling seasons, which are
further enhanced by the shaded regions. The multidimensional scaling was based on
Bray-Curtis similarity index and biomass (fourth root transformed).
27
The effect of seasonal establishment on the rate of succession can however
vary with the stage in the succession. Species that colonise during summer
are able to occupy all space thanks to rapid growth in the warmer water and
can thus inhibit settlement by later species. Thus, the succession is initially
rapid on substrate available in summer, due to many colonisers and rapid
growth (total biomass on three-month-old plates was highest on July-plates),
but as the established species inhibits new colonisers the rate decreases.
Even relatively stable communities can develop as species are excluded from
the communities (paper IV). On the other hand, succession will initially be
slow on substrate available in the colder winter months, when few species
can colonise and the growth is slow (total biomass on three-month-old plates
were lowest on January and October-plates). Due to the slow growth, space
remains available for colonisation by more species in spring and early
summer, thereby increasing the rate of succession.
The type of species interactions in the succession could further influence the
succession rate. A winter recruiting species with prolific settlement may gain
a head start on competitors settling later. Thus, species able to colonise
during the colder months may be able to obtain a firm hold on free space,
thereby inhibiting later species and slowing the succession. For example, the
annual brown alga Pilayella littoralis successfully colonised substrate
available in late autumn and seemed able to reduce new colonisation during
at least 30 months (paper IV). The succession may also be accelerated if
early species or species able to invade despite the presence of other species
facilitate colonisation of new species. This was observed for Mytilus edulis
which seemed to facilitate establishment by Furcellaria lumbricalis, the
dominant alga in the surrounding communities (paper IV).
28
Acknowledgements
I am very grateful to all family, friends and colleagues who have supported
me in this thesis work. I am especially grateful to my supervisors, Hans
Kautsky and Lena Kautsky, whose enthusiasm and support have inspired and
made this thesis possible. Special thanks also to Klemens Eriksson, Antonia
Sandman, Sofia Wikström, Anders Wallin, Nils Kautsky, Pia Pettersson and
my co-author Torleif Malm for inspiring discussions and helpful comments.
Especially, Sonja Råberg who has shared the ups and downs of these years. I
am also very grateful to Lars Gustafsson who rescued the contents from my
crashed computer – twice!
Many thanks also to all who have helped in the field and lab work. My
fellow shipmates on “Corfven af Calmar” on the bridge adventure, Ellen
Schagerström and Johan Storck. All colleagues and friends who have kept
me company during the many dives in the settlement and succession studies,
Erica Johannesson, Sonja Råberg, Sofia Wikström, Pia Pettersson, Antonia
Sandman and Anders Wallin. The always helpful staff at Askö, who carried
plates and watched my bubbles in all weather, Susann Ericsson, Mattias
Murphy, Stefan Andersson, Eddie Eriksson and Mikael Karlsson.
And most importantly, thank you, mum and dad for laying the foundations
for my great interest and fascination with the sea that lead to this thesis.
Thank you Isac, for your never-ending support, despite long field periods in
the best sailing season and never-kept promises of “next summer will be
better”. And of course for being an emergency dive buddy on occasion, even
in the middle of winter, and for help with design and construction of
experimental equipment. I am grateful for the support from my family, both
old and new, that has kept me going through the rough times. Thank you,
mum, dad, Kenneth, Isac, Annette, Harry, Marie, Bo, Niklas and Tobias.
Financial support was provided by the Alice and Lars Silén fund, the C. F.
Lundström fund, Stockholm Marine Research Centre (SMF), C. A. Öberg
fund and the Swedish Energy Agency.
29
References
Allen JA, Cook M, Jackson DJ, Preston S, Worth EM (1976) Observations on the
rate of production and mechanical properties of the byssus threads of Mytilus
edulis L. Journal of Molluscan Studies 42: 279-289
Amsler CD, Searles RB (1980) Vertical distribution of seaweed spores in a water
column offshore of North Carolina. Journal of Phycology 16: 617-619
Ang PJ (1985) Studies on the recruitment of Sargassum spp. (Fucales: Phaeophyta)
in Balibago, Calatagan, Phillippines. Journal of Experimental Marine Biology
and Ecology 91: 293-301
Benedetti-Cecchi L, Cinelli F (1993) Early patterns of algal succession in a
midlittoral community of the Mediterranean sea: a multifactorial experiment.
Journal of Experimental Marine Biology and Ecology 169: 15-31
Berger R, Henriksson E, Kautsky L, Malm T (2003) Effects of filamentous algae
and deposited matter on the survival of Fucus vesiculosus L. germlings in the
Baltic Sea. Aquatic Ecology 37: 1-11
Bergström L, Bruno E, Eklund B, Kautsky L (2003) Reproductive strategies of
Ceramium tenuicorne near its inner limit in the brackish Baltic Sea. Botanica
Marina 46: 125-131
Breitburg DL (1985) Development of a subtidal epibenthic community: factors
affecting species composition and the mechanisms of succession. Oecologia 65:
173-184
Caffey H (1985) Spatial and temporal variation in settlement and recruitment of
intertidal barnacles. Ecological Monographs 55: 313-332
Chalmer PN (1982) Settlement patterns of species in a marine fouling community
and some mechanims of succession. Journal of Experimental Marine Biology
and Ecology 58: 73-85
Chapman MG (2002) Early colonization of shallow subtidal boulders in two
habitats. Journal of Experimental Marine Biology and Ecology 275: 95-116
Chapman MG (2003) Paucity of mobile species on constructed seawalls: effects of
urbanization on biodiversity. Marine Ecology Progress Series 264: 21-29
Christie H (1985) Ecological monitoring strategy with special reference to a rocky
subtidal programme. Marine Pollution Bulletin 16: 232-235
Clements FE (1916) Plant succession: an analysis of the development of vegetation.
Carnegie institute, Washington, D.C.
Coma R, Ribes M, Gili J-M, Zabala M (2000) Seasonality in coastal bentic
ecosystems. Tree 15: 448-453
Connell JH (1961) The influence of interspecific competition and other factors on
the distribution of the barnacle Chthamalus Stellatus. Ecology 42: 710-723
Connell JH (1985) The consequences of variation in initial settlement vs. postsettlement mortality in rocky intertidal communities. Journal of Experimental
Marine Biology and Ecology 93: 11-45
Connell JH, Slayter RO (1977) Mechanisms of succession in natural communities
and their role in community stability and organization. The American Naturalist
111: 1119-1144
Connell JH, Sousa WP (1983) On the evidence needed to judge ecological stability
or persistence. The American Naturalist 121: 789-824
30
Connell SD, Glasby TM (1999) Do urban structures influence local abundance and
diversity of subtidal epibiota? A case study from Sydney Harbour, Australia.
Marine Environmental Research 47: 373-387
Dayton PK (1971) Competition, disturbance, and community organization: the
provision and subsequent utilization of space in a rocky intertidal community.
Ecological Monographs 41: 351-389
Dayton PK (1973) Dispersion, dispersal and persistence of the annual intertidal alga
Postelsia palmaeformis ruprecht. Ecology 54: 433-438
Dayton PK, Currie V, Gerrodette T, Keller BD, Rosenthal R, Tresca DV (1984)
Patch dynamics and stability of some California kelp communities. Ecological
Monographs 54: 253-289
Dean TA, Hurd LE (1980) Development in an estuarine fouling community: the
influence of early colonists on later arrivals. Oecologia 46: 295-301
Deysher L, Norton TA (2003) Dispersal and colonization in Sargassum muticum
(Yendo) Fensholt. Journal of Experimental Marine Biology and Ecology 56 (23): 179-195
Dietrich G, Schott F (1974) Eisverhältnisse. In Meereskunde der Ostesee. Magaard
L, Rheinheimer G (eds), Springer-Verlag, Berlin. p. 61-66.
Drury WH, Nisbet ICT (1973) Succession. Journal of Arnold Arboretum Harvard
University 54: 331-368
Dürr S, Wahl M (2004) Isolated and combined impacts of blue mussels (Mytilus
edulis) and barnacles (Balanus improvisus) on structure and diversity of a
fouling community. Journal of Experimental Marine Biology and Ecology 306:
181-195
Emerson SE, Zedler JB (1978) Recolonization of intertidal algae: an experimental
study. Marine Biology 44: 315-324
Enderlein P, Wahl M (2004) Dominance of blue mussels versus consumer-mediated
enhancement of benthic diversity. Journal of Sea Research 51: 145-155
Engkvist R, Malm T, Nilsson J (2004) Interaction between isopod grazing and wave
action: a structuring force in macroalgal communities in the southern Baltic Sea.
Aquatic Ecology 38: 403-413
Engkvist R, Malm T, Tobiasson S (2000) Density dependent grazing effects of the
isopod Idotea baltica Pallas on Fucus vesiculosus L in the Baltic Sea. Aquatic
Ecology 34: 253-260
Fager EW (1971) Pattern in the development of a marine community. Limnology
and Oceanography 16: 241-253
Farrell TM (1991) Models and mechanisms of succession: an example from a rocky
intertidal community. Ecological Monographs 61: 95-113
Fletcher RL, Callow ME (1992) The settlement, attachment and establishment of
marine algal spores. British Phycological Journal 27: 303-329
Foster MS (1975) Algal succession in a Macrocystis pyrifera forest. Marine Biology
32: 313-329
Glasby TM (1999) Interactive effects of shading and proximity to the seafloor on the
development of subtidal epibiotic assemblages. Marine Ecology Progress Series
190: 113-124
Greene CH, Schoener A (1982) Succession on marine hard substrata: A fixed
lottery. Oecologia 55: 289-297
Haderlie EC (1969) Marine fouling and boring organisms in Monterey Harbor. II.
Second year of investigation. The Veliger 12: 182-192
31
Harlin MM, Lindbergh JM (1977) Selection of substrata by seaweeds: optimal
surface relief. Marine Biology 40: 33-40
Harms J, Anger K (1983) Seasonal, annual, and spacial variation in the development
of hard bottom communities. Helgolander Meeresuntersuchungen 36: 137-150
Hartnoll RG (1983) Substratum. In: Sublittoral ecology. The ecology of the shallow
sublittoral benthos. Earll, R. and Erwin, D.G. (eds). Clarendon press. Oxford. p.
97-124.
Hawkins SJ (1981) The influence of season and barnacles on the algal
recolonization of Patella vulgata exclusion areas Journal of the marine
Biological Association of U. K. 61: 1-15
Hawkins SJ, Hartnoll RG (1982) Settlement patterns of Semibalanus balanoides (L.)
in the Isle of Man (1977-1981). Journal of Experimental Marine Biology and
Ecology 62: 271-283
Hoffmann AJ (1987) The arrival of seaweed propagules at the shore: a review.
Botanica Marina 30: 151-165
Hoffmann AJ, Ugarte R (1985) The arrival of propagules of marine macroalgae in
the intertidal zone. Journal of Experimental Marine Biology and Ecology 92:
83-95
Horn H (1974) The ecology of secondary succession. Annual Review of Ecology
and Systematics 5: 25-37
Hosomi A (1977) Ecological studies of the mussel, Mytilus galloprovincialis
(LAMARCK), I. The fluctuation of its coverage. Japanese Journal of Ecology
27: 311-318
Hruby T, Norton TA (1979) Algal colonization on rocky shores in the Firth of
Clyde. Journal of Ecology 67: 65-77
Hutchinson N, Williams GA (2001) Spatio-temporal variation in recruitment on a
seasonal, tropical rocky shore: the importance of local versus non-local
processes. Marine Ecology Progress Series 215: 57-68
Jansson A-M, Kautsky N (1977) Quantitative survey of hard bottom communities in
a Baltic archipelago. In: Keegan B, Ceidigh P, Boaden P (eds) Biology of
Benthtic Organisms. Pergamon Press London and New York, pp 359-366
Jernakoff P (1985) Temporal and small-scale spatial variability of algal abundance
on an intertidal rocky shore. Botanica Marina 28: 145-154
Johansson G, Eriksson BK (manus.) Reattachment of macroalgal fragments enables
population persistence in a suboptimal environment. In PhD thesis, Faculty of
Science and Technology 759, Uppsala University, Sweden.
Johansson G (2002) Factors affecting the distribution of rocky-shore macroalgae on
the Swedish coast. Morphological, physiological, reproductive and genetic
aspects. PhD thesis, Faculty of Science and Technology 759, Uppsala
University, Sweden.
Kain JM (1975) Algal recolonization of some cleared subtidal areas. Journal of
Ecology 63: 739-765
Kautsky H (1989) Quantitative distribution of plant and animal communities of the
phytobenthic zone in the Baltic Sea. Contributions from the Askö laboratory 35:
1-80
Kautsky H (1995) Quantitative distribution of sublittoral plant and animal
communities along the Baltic Sea gradient. In: Eleftheriou A (ed) 28th EMBS
Symposium: Biology and Ecology of Shallow Coastal Waters. Olsen & Olsen,
Denmark, pp 23-30
32
Kautsky H, Maarel Evd (1990) Multivariate approaches to the variation in
phytobenthic communities and environmental vectors in the Baltic Sea. Marine
Ecology Progress Series 60: 169-184
Kautsky U, Wallentinus I, Kautsky N (1984) Spring bloom dynamics of an epilithic
microphytobenthic community in the northern Baltic proper. Ophelia 3: 89-99
Keough MJ (1984) Effects of patch size on the abundance of sessile marine
invertebrates. Ecology 65: 423-437
Keough MJ, Downes BJ (1982) Recruitment of marine invertebrates: the role of
active larval choices and early mortality. Oecologia 54: 348-352
Kiirikki M, Lehvo A (1997) Life strategies of filamentous algae in the northern
Baltic Proper. Sarsia 82: 259-267
Kim JH (1997) The role of herbivory, and direct and indirect interactions, in algal
succession. Journal of Experimental Marine Biology and Ecology 217: 119-135
Littorin B, Gilek M (1999) A photographic study of the recolonization of cleared
patches in a dense population of Mytilus edulis in the northern Baltic proper.
Hydrobiologia 393: 211-219
Lotze HK, Schramm W, Schories D, Worm B (1999) Control of macroalgal blooms
at early developmental stages: Pilayella littoralis versus Enteromorpha spp.
Oecologia 119: 46-54
Lotze HK, Worm B (2000) Variable and complementary effects of herbivores on
different life stages of bloom-forming macroalgae. Marine Ecology Progress
Series 200: 167-175
Lubchenco J (1983) Littorina and Fucus: Effects of herbivores, substratum
heterogeneity, and plant escapes during succession. Ecology 64: 1116-1123
Lubchenco J, Menge BA (1978) Community development and persistence in a low
rocky intertidal zone. Ecological Monographs 48: 67-94
Magaard L (1974) Wasserstandsschwankungen und seegang. In Magaard L,
Rheinheimer G (eds). Meereskunde der Ostesee. Springer-Verlag, Berlin. p. 6776.
Malm T, Engkvist R, Kautsky L (1999) Grazing effects of two freshwater snails on
juvenile Fucus vesiculosus in the Baltic Sea. Marine Ecology Progress Series
188: 63-71
Malm T, Kautsky L, Claesson T (2003) The density and survival of Fucus
vesiculosus L. (Fucales, Phaeophyta) on different bedrock types on a Baltic Sea
moraine coast. Botanica Marina 46: 256-262
Malm T, Engkvist R, Kautsky L (manus.) Fucus vesiculosus L. and Fucus serratus
L. (Fucales Phaeophyceae) as food and habitat for Idotea baltica
(Pallas)(Isopoda, Crustacea) in the Baltic Sea. In Ph.D. thesis Distribution
patterns and ecology of Fucus serratus L. and Fucus vesiculosus L. in the Baltic
Sea. Department of Botany, Stockholm University, Stockholm.
Mathieson AC (1966) Morphological studies of the marine brown alga Taonia
lennebackerae Farlow ex. J. Agardh. I. Sporophytes, abnormal gametophytes
and vegetative reproduction. . Nova Hedwigia 12: 65-79
McCook LJ, Chapman ARO (1991) Community succession following massive icescour on an exposed rocky shore: effects of Fucus canopy algae and mussels
during late succession. Journal of Experimental Marine Biology and Ecology
154: 137-169
33
McCook LJ, Chapman ARO (1992) Vegetative regeneration of Fucus rockweed
canopy as a mechanism of secondary succession on an exposed rocky shore.
Botanica Marina 35: 35-46
McGuinness KA, Underwood AJ (1986) Habitat structure and the nature of
communities on intertidal boulders. Journal of Experimental Marine Biology
and Ecology 104: 97-123
Menge BA (1976) Organization of the New England rocky intertidal community:
role of predation, competition, and environmental heterogenity. Journal of
Experimental Marine Biology and Ecology 46: 355-393
Menge BA (1991) Relative importance of recruitment and other causes of variation
in rocky intertidal community structure. Journal of Experimental Marine
Biology and Ecology 146: 69-100
Mook DH (1981) Effects of disturbance and initial settlement on fouling community
structure. Ecology 62: 522-526
Moran PJ, Grant TR (1989a) The effects of industrial pollution on the development
and succession of marine fouling communities. P.S.Z.N.I: Marine Ecology 10:
231-246
Moran PJ, Grant TR (1989b) The effects of industrial pollution the development and
succession of marine fouling communities. II. Multivariate analysis of
succession. Marine Ecology 10: 247-261
Norton TA, Fetter R (1981) The settlement of Sargassum muticum propagules in
stationary and flowing water. Journal of the marine Biological Association of U.
K. 61: 929-940
Norton TA (1985) The zonation of seaweeds on rocky shores. In: The ecology of
rocky coasts. PG Moore & R Seed (eds). Hodder and Stoughton, London. p. 722.
Odum EP (1969) The strategy of ecosystem development. Science 164: 262-270
Osman RW (1977) The establishment and development of a marine epifaunal
community. Ecological Monographs 47: 37-63
Raberg S; Berger-Joensson R, Bjoern A, Graneli E, Kautsky L (2005) Effects of
Pilayella littoralis on Fucus vesiculosus recruitment: implications for
community composition. Marine Ecology Progress Series 289: 131-139.
Santelices B (1990) Patterns of reproduction, dispersal and recruitment in seaweeds.
Oceanography and Marine Biology. An Annual Review 28: 177-276
Scheer BT (1945) The development of marine fouling communities. Biological
Bulletin 89: 103-121
Serisawa Y, Taino S, Ohno M, Aruga Y (1998) Succession of seaweeds on
experimental plates immersed during different seasons in Tosa Bay, Japan.
Botanica Marina 41: 321-328
Serrão EA, Kautsky L, Lifvergren T, Brawley SH (1997) Gamete dispersal and prerecruitment mortality in Baltic Fucus vesiculosus. Phycologia (Suppl.) 36: 101102
Sjöberg B, Larsson U (1995) Hydrografi och näringsämnen. Östersjö 95, årsrapport.
Stockholms Marina Forskningscentrum. Stockholm.
Skoog G (1978a) Aspects on the biology and ecology of Theodoxus fluviatilis (L.)
and Lymnea peregra (O.F.Müller)(Gastropoda) in the northern Baltic.
Department of Zoology and the Askö Laboratory, Stockholm
34
Skoog G (1978b) Influence of natural food items on growth and egg production in
brackish water populations of Lymnea peregra and Theodoxus fluviatilis. Oikos
31: 340-348
Sousa WP (1979) Experimental investigations of disturbance and ecological
succession in a rocky intertidal algal community. Ecological Monographs 49:
227-254
Sousa WP (1984) Intertidal mosaics: patch size, propagule availability, and spatially
variable patterns of succession. Ecology 65: 1918-1935
Sousa WP, Connell JH (1992) Grazing and succession in marine algae. In: John D,
Hawkins S, Price J (eds) Plant-Animal Interactions in the Marine Benthos.
Systematics Association Special Volume, Clarendon Press, Oxford, pp 425-441
Sutherland JP (1974) Multiple stable points in natural communities. American
Naturalist 108: 859-873
Sutherland JP, Karlson RH (1977) Development and stability of the fouling
community at Beaufort, North Carolina. Ecological Monographs 47: 425-446
Svedelius N (1901) Studier öfver Östersjöns hafsalgflora. PhD thesis, Uppsala
University, Sweden (in Swedish)
Tobiasson S, Lundgren F, Sjölin A, Wickström K (2001) Blekingekustens
Vattenvårdsförbund och Vattenvårdsförbundet för västra Hanöbukten.
Årsrapport. Kalmar Högskola. Rapport 2001:4. pp 73.
Turner MG, Baker WL, Peterson CJ, Peet RK (1998) Factors influencing
succession: Lessons from large, infrequent natural disturbances. Ecosystems 1:
511-523
Turner T (1983a) Complexity of early and middle successional stages in a rocky
intertidal surfgrass community. Oecologia 60: 56-65
Turner T (1983b) Facilitation as a successional mechanism in a rocky intertidal
community. The American Naturalist 121: 729-738
Underwood AJ, Anderson MJ (1994) Seasonal and temporal aspects of recruitment
and succession in an intertidal estuarine fouling assemblage. J. Mar. Biol.
Assoc. U. K. 74: 563-584
Vadas Sr RL, Johnson S, Norton TA (1992) Recruitment and mortality of early postsettlement stages of benthic algae. British Phycological Journal 27: 331-351
Waern M (1952) Rocky-shore algae in the Öregrund archipelago. In Acta
Phytogeogr. Suec. 30. Almqvist & Wiksells boktryckeri AB, Uppsala.
Wahl M (2001) Small scale variability of benthic assemblages: biogenic
neighborhood effects. Journal of Experimental Marine Biology and Ecology
258: 101-114
Wallentinus I (1979) Environmental influences on benthic macrovegetation in the
Trosa - Askö area, northern Baltic proper II. The ecology of macroalgae and
submersed phanerogams. Contributions from the Askö laboratory 25: 1-210
Westring, G., 1993. Isförhållanden i svenska farvatten under normalperioden 1961–
1990. Report no. SMHI Oceanografi 59. Swedish Meterological and
Hydrological Institute, Norrköping, 68 pp.
Westring, G., 1995. Isförhållanden utmed Sveriges kust. Isstatistik från svenska
farleder och farvatten under normalperioderna 1931–1961 och 1961–1990.
Report no. SMHI Oceanografi 63. Swedish Meterological and Hydrological
Institute, Norrköping, 24 ppWilson OT (1925) Some experimental observations
of marine algal successions. Ecology 6: 303-311
35
Worm B, Lotze HK, Boström C, Engkvist R, Labanauskas V, Sommer U (1999)
Marine diversity shift linked to interactions among grazers, nutrients and
propagule banks. Marine Ecology Progress Series 185: 309-314
Worm B, Lotze HK, Sommer U (2001) Algal propagule banks modify competition,
consumer and resource control on Baltic rocky shores. Oecologia 128: 281-293
Young GA (1985) Byssus-thread formation by the mussel Mytilus edulis: effects of
environmental factors. Marine Ecology Progress Series 24: 261-271
Zechman FW, Mathieson AC (1985) The distribution of seaweed propagules in
estuarine, coastal and offshore waters of New Hampshire, U.S.A. Botanica
Marina 28: 283-294
36
Swedish summary / svensk sammanfattning
Årstidens betydelse för rekrytering och succession i
Östersjöns grunda hårdbottensamhällen.
De grunda hårda bottnarna längs våra kuster sjuder av liv. I dessa
hårdbottensamhällen bildar makroalger och fastsittande djur, som musslor
och havstulpaner, den tredimensionella struktur som fiskar och andra djur
lever i. Naturliga störningar i form av t ex isskrap och stormar öppnar upp
fria ytor genom att slita bort växter och djur. På dessa nya ytor sker en
nyrekrytering av arter och med tiden utvecklas nya samhällen. Detta förlopp
kallas succession. Kustområden påverkas både indirekt via t ex utsläpp från
avrinningsområdet och direkt genom utnyttjande av kustens resurser i form
av t ex fiske och rekreation. Miljöövervakning av hårdbottensamhällen
bedrivs för att undersöka hälsotillståndet och varna för förändringar. För att
kunna identifiera eventuella förändringar orsakade av mänsklig påverkan
måste man dock känna till den naturliga variationen i samhällena.
Jag har undersökt hur rekrytering och succession i Östersjöns grunda
hårdbottensamhällen varierar beroende på vilken tid på året en yta blir
tillgänglig för kolonisation. Alger koloniserar vanligen nya ytor genom att
producera sporer eller liknande förökningskroppar som sprids med
vattenströmmar. Fastsittande djur har frisimmande larver som även de främst
sprids med vattenströmmarna. Genom att placera ut plastskivor på botten
varje månad under två år har jag beskrivit när på året olika arter har
möjlighet att kolonisera nya ytor. Generellt koloniserade en given makroalg
eller ett fastsittande djur nya ytor bara under en relativt kort period varje år
medan några få arter kunde kolonisera nya ytor året runt. De flesta arter
rekryterade under sommar och tidig höst. Min studie visade endast små
skillnader i rekrytering mellan år och mellan olika platser i skärgården,
vilket indikerar att den återkommande årstidsstyrda rekryteringen är relativt
lätt att förutsäga. Successionen har studerats genom att placera ut
granitplattor under olika årstider och följa utvecklingen av nya samhällen
under två år. Artsammansättningen bestämdes initialt av vilka arter som
råkade ha en rekrytering som sammanföll med plattutsättningarna och
därmed hade möjlighet att kolonisera plattorna. Resultatet blev att
successionen på plattorna började med olika arter. Arter påverkar varandra
och kan genom att t ex ockupera all fri yta hindra andra arter från att
kolonisera. Arter kan också gynnas av att en annan art är där före dem.
Successionen på plattorna såg olika ut beroende på när plattorna placerades
ut på botten. Med tiden blev plattsamhällena dock allt mer lika de samhällen
som växte på omgivande botten. Detta tyder på att det finns ett förutsägbart
slutsamhälle som bestäms av rådande omvärldsfaktorer. Vilka arter som
37
koloniserade först verkade mest påverka hur snabbt slutsamhället nåddes. På
plattor som placerades på botten i oktober koloniserade en ettårig fintrådig
brunalg, Pilayella littoralis i stora mängder. Den verkade därefter i stor
utsträckning kunna hindra andra från att kolonisera vilket bromsade
successionen på dessa plattor. Plattor som sjösattes i april, innan
högsäsongen i rekrytering och tillväxt, kunde koloniseras av de flesta arterna
och fick snabbt ett samhälle som var mycket likt de omgivande
hällsamhällena. Resultaten visar att återhämtningen efter en störning i dessa
grunda hårdbottensamhällen påverkas av när på året störningen sker.
Nya ytor blir också tillgängliga när kustområden exploateras. När t ex
hamnar, broar och pirar byggs tillförs havet nya konstgjorda hårdbottnar,
som snabbt koloniseras av både flora och fauna. I en fältstudie har jag
studerat samhällen och succession på artificiellt substrat i form av bropelare
renoverade under en tioårsperiod. Ytan som bropelarna erbjöd var vertikal
och mycket slät, därför jämfördes samhällena på bropelarna med samhällen
på vertikala naturliga substrat i närheten av bron. Bropelarsamhällena hade
färre arter, speciellt fleråriga algarter, än det naturliga substratet, troligen
berodde detta på pelarens släta yta. Detta visar att nya konstgjorda hårda
bottnar inte är jämförbara med naturliga substrat och i detta fall tillförde de
ingen ökad mångfald i området.
I en annan fältstudie har jag undersökt hur rekrytering och samhällstruktur
påverkas av lutningen på botten. Artificiella konstruktioner som t ex
pontoner och bryggor samt även naturliga substrat som t ex block erbjuder
ytor som befinner sig ovanför den omgivande botten. Därför undersökte jag
också rekryteringen på olika höjd över botten. Det var skillnader i
rekrytering beroende på både lutning och höjd över botten. Flera arter, bl a
norra Östersjöns enda stora tångart blåstången Fucus vesiculosus, hade
betydlig sämre rekrytering bara 40 cm ovanför botten jämfört med nära
botten. Blåstångens rekrytering försämrades också gradvis med ökad lutning
av substratet och på vertikalytor eller överhäng observerades bara enstaka
groddar. Rekryteringen av många arter förändrades mycket just mellan 60º
och 90º. Ett liknade mönster iakttogs när naturliga samhällen på häll med
samma lutning undersöktes. På hällar med lutning mellan 0º och 60º
dominerade alger medan vertikala hällar dominerades av fastsittande djur,
främst blåmussla Mytilus edulis och havstulpan Balanus improvisus. Det
observerade mönstret i rekrytering som återspeglades i de naturliga
samhällena tyder på att detaljer i slutsamhällenas struktur kan bestämmas
redan vid rekryteringsstadiet.
Jag har visat att successionen varierar beroende på när under året en yta
frigörs. Vilka arter som koloniserar först beror på tidpunkten när ytan
friläggs och detta kan påverka hur lång tid det tar för samhället att återhämta
sig efter en störning. De små skillnaderna i rekrytering mellan år och plats
38
samt ett slutsamhälle som främst verkade bestämmas av rådande
omvärldsfaktorer tyder på successionen i Östersjöns grunda hårdbottensamhällen är relativt förutsägbar. Mina resultat skulle kunna hjälpa oss att
tolka naturliga variationer i samhällen och särskilja mänsklig påverkan ur
miljöövervakningssynpunkt. Det skulle kanske också kunna hjälpa oss att
förkorta samhällens återhämtningstid efter planerade störningar, t ex
muddring eller nybyggnation, genom att avsluta störningar vid lämplig
tidpunkt på året.
39