Content uploaded by Ohad Peleg
Author content
All content in this area was uploaded by Ohad Peleg on Dec 31, 2017
Content may be subject to copyright.
RESEARCH ARTICLE
Alien turf: Overfishing, overgrazing and invader domination in
south‐eastern Levant reef ecosystems
Gil Rilov
1,2
|Ohad Peleg
1,2
|Erez Yeruham
1,2
|Tal Garval
1,2
|Ania Vichik
1
|Ofrat Raveh
1
1
National Institute of Oceanography, Israel
Oceanographic and Limnological Research
(ILOR), Haifa, Israel
2
Marine Biology Department, Charney School
of Marine Science, University of Haifa, Haifa,
Israel
Correspondence
Gil Rilov, National Institute of Oceanography,
Israel Oceanographic and Limnological
Research (ILOR), Haifa 3108001, Israel.
Email: rilovg@ocean.org.il
Funding information
Israel Ports Company; Goldman Philanthropic
Partnerships; European Commission, Grant/
Award Number: 249147; Israel Science Foun-
dation, Grant/Award Number: 1217/10; Marie
Curie Reintegration Programme under the
European Union's Seventh Framework, Grant/
Award Number: 249147
Abstract
1. Coastal reefs are highly diverse marine ecosystems that in many regions suffer today from
growing pressures by human activities. Among the most highly‐stressed are those found in
the Levantine basin (south‐eastern Mediterranean Sea). The Levant represents the trailing‐
edge of distribution of native species where they are exposed to the most extreme tempera-
ture and salinity conditions, and the region is also fast‐warming and exposed to a great many
alien species and strong fishing pressure. In this study, the ecological state of reefs in the
south‐eastern Levant was assessed quantitatively (including inside a small marine reserve)
using current, extensive, survey data with reference to anecdotal historical information on
their more pristine past.
2. The results of very extensive subtidal community surveys that were conducted in north Israel
indicate that reefs in this area are currently dominated by turf‐forming algae and aliens, and
sustain low numbers of top predators. Specifically, it was found that on these Levant reefs:
(1) commercial species represent a very small part of the fish assemblage (except inside the
reserve); (2) alien species constitute a considerable portion (23–44%) of the fish assemblage
(including in the reserve) and 95–99% of epi‐benthic molluscs, including inside the marine
reserve; and (3) turf barrens are the dominant substrate cover, while cover of native brown
algae canopy is limited to small patches occurring only during winter and spring.
3. These findings suggest that the Levant reefs have been highly transformed by overfishing and
alien invasions, and probably also climate change, and that even well managed marine reserves
had little effect on alien species presence. From a biogeographic‐conservation perspective, as
both warming and bioinvasions continue in the Mediterranean, it is expected that this
degraded reef state will gradually advance westward. Alleviating fishing pressure with marine
reserves might make the reefs more resilient to these regional pressures, but alien invaders will
remain a dominant feature in the system. Therefore, a more realistic conservation target might
be the preservation or restoration of ecosystem functions rather than the original native
biodiversity.
KEYWORDS
alien species, biodiversity, climate change, coastal, fishing, marine protected areas
1|INTRODUCTION
In the Anthropocene (Steffen, Crutzen, & McNeill, 2007), marine
ecosystems are increasingly exposed to human‐mediated stresses,
among them pollution, habitat destruction, overexploitation, climate
change and bioinvasions (Gattuso et al., 2015; Halpern et al., 2015;
Jackson, 2008; Poloczanska et al., 2013). This pressure on the marine
environment results in accelerated rates of alteration of biological
communities expressed as local changes in biodiversity, food webs
and ecosystem functions, as well as the expansion or retraction of
species distributions at larger scales. Among the most biologically‐
diverse and fragile marine ecosystems are coastal reefs (both coral
and rocky), and in recent years, increasing numbers of reefs globally
have been transforming from one structural and functional state to a
Received: 4 September 2016 Revised: 5 October 2017 Accepted: 8 October 2017
DOI: 10.1002/aqc.2862
Aquatic Conserv: Mar Freshw Ecosyst. 2017;1–19. Copyright © 2017 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/aqc 1
different, usually more degraded, state. Examples of state shifts include
coral reefs overgrown by macroalgae owing to extreme eutro-
phication and/or the removal of grazers, or exposure to extreme heat
waves (e.g. during El Niño) that cause coral bleaching (Aronson et al.,
2003), kelp forests losing their main habitat formers (kelps) because
of eutrophication, the removal of top predators (through cascading tro-
phic effects) or heat waves (Steneck et al., 2002; Wernberg et al., 2013),
and rocky reefs overgrazed by sea urchins as a result of the removal of
their predators by fishing (Guidetti, 2006; Sala, Boudouresque, &
Harmelin‐Vivien, 1998), or by tropical (often alien) herbivorous fish
moving into warming temperate areas (Vergés, Steinberg, et al., 2014;
Vergés et al., 2016). This study explores the evidence for a fundamental
state shift on south‐eastern Mediterranean reefs.
The Mediterranean Sea –the largest enclosed sea on the planet –
has been exposed to extensive anthropogenic stressors for millennia
(Coll et al., 2012). Besides local stressors such as pollution and
overexploitation, it is also fast‐warming (Lejeusne, Chevaldonné,
Pergent‐Martini, Boudouresque, & Pérez, 2010; Shaltout & Omstedt,
2014), and is heavily invaded by (mostly) thermophilic species
(Katsanevakis et al., 2014; Rilov & Galil, 2009). However, studies on
the impacts of these stresses on Mediterranean reefs have mainly
focused on the western basin where in many places overfishing has
transformed the rich benthos to urchin‐dominated barrens (Agnetta
et al., 2015; Sala & Zabala, 1996), and alien species, mainly highly
invasive macroalgae, have been taking over the benthos by
outcompeting native species (Ferrer, Garreta, & Ribera, 1997; Piazzi,
Ceccherelli, & Cinelli, 2001; Wright, McKenzie, & Gribben, 2007). In
the past two decades, the western basin also suffered from two
extreme heat waves causing mass mortalities of sessile invertebrates
(Garrabou et al., 2009; Marbà, Jorda, Agusti, Girard, & Duarte, 2015).
The eastern basin, however, is perhaps even more prone to rapid eco-
system transformation than the western basin. Its coastal water condi-
tions are naturally more extreme, being hottest and saltiest in the
Mediterranean (Coll et al., 2010). It is also warming faster than the
western basin (Shaltout & Omstedt, 2014; Sisma‐Ventura, Yam, &
Shemesh, 2014), with a sea surface temperature (SST) increase of
2–3°C in the past three decades (Rilov, 2016). The Israeli coast in
the south‐eastern Levant is also the first stop for most Indo‐Pacific
invaders arriving directly from the Rea Sea through the Suez Canal,
making it the ‘hottest hotspot’for marine alien species in the Mediter-
ranean and probably globally (Edelist, Rilov, Golani, Carleton, &
Spanier, 2013; Katsanevakis et al., 2014). The Levantine basin repre-
sents the south‐eastern, warm (‘trailing’) edge of distribution of most
Mediterranean and Atlanto‐Mediterranean species, and thus where
species sensitive to warming might be expected to disappear first.
Trawl data analysis has recently shown a major shift of soft‐bottom
fish communities on the Israeli coast over the past two decades
(Edelist, Rilov, et al., 2013), probably a result of a combination of
overfishing, bioinvasions and ocean warming. Because these stressors
also pose a threat to the integrity of Levant reefs, considerable ecolog-
ical transformations are expected in this system too.
Recently, a large group of ecologists executed a major collabora-
tive, basin‐wide, very‐detailed nearshore reef community survey that
included fish, macroalgae and invertebrates (Sala et al., 2012). This
effort was set to establish a baseline for Mediterranean reef health
status, and focused on the effects of protection level on reef commu-
nities (Sala et al., 2012); however, sampling in the Levant during that
study was extremely minimal and did not include the south‐eastern
corner (i.e. Syria, Lebanon or Israel) –the Mediterranean's warmest
and most highly‐invaded region. Sala et al. have identified three alter-
native community states for Mediterranean reefs: (1) reefs that are
dominated by non‐canopy algae and have high fish biomass; (2) reefs
with abundant native algal canopy (mostly brown algae) and suspen-
sion feeders, that have lower fish biomass; and (3) reefs with extensive
barrens or turf algal cover and low fish biomass.
Although comprehensive reef community surveys were absent in
the south‐eastern Levant (until 2010, see Rilov, 2016), several studies
over the past few decades have shown varied evidence of reef
community transformation in the region. On the Israeli and Lebanese
coasts, a few local surveys of fish assemblages have shown the
domination of fish biomass by non‐indigenous species, mostly two
highly abundant, herbivorous, Indo‐Pacific rabbitfish species (Siganus
rivulatus and Siganus luridus, Goren & Galil, 2001; Harmelin‐Vivien,
Bitar, Harmelin, & Monestiez, 2005). Recent gut and isotope analysis
of native and invasive reef fish in Lebanon suggested a shift in the
local food web (Fanelli, Azzurro, Bariche, Cartes, & Maynou, 2015).
On the Israeli coast, population collapses of many native, non‐
harvested, reef molluscs and two sea urchins have recently been
reported (Rilov, 2016). Experimental work has shown that the almost
complete disappearance of sea urchins could be explained by ocean
warming, as the urchins die when SST exceeds 30.5°C (Yeruham, Rilov,
Shpigel, & Abelson, 2015), which has occurred every summer since
1998 (Rilov, 2016). However, competition for food with the invasive
rabbitfish might have also contributed to the urchins’collapse
(Yeruham, 2013). In the north‐western part of the Turkish Levant
coast, field experiments have indeed shown that the rabbitfish prevent
the development of brown algae meadows by overgrazing (Sala,
Kizilkaya, Yildirim, & Ballesteros, 2011). Currently, there are no studies
assessing the ecological impacts of fishing pressure on reefs within the
Levant. A recent assessment by Edelist, Scheinin, et al. (2013)
suggested an increase (mostly in the last decade) in unregulated,
poorly‐reported, nearshore artisanal and recreational fishing in Israel;
sectors that mostly fish over rocky reefs. More than 70 000 Israelis,
with increasingly sophisticated gear, are engaged in recreational fishing
activity, and the take was estimated at over 15 500 tons; 20% of the
total estimated catch for that coast. All these findings suggest that
bioinvasions, ocean warming, and possibly overfishing, may be acting
together to drastically transform Levant reef ecosystems.
This study presents results from two reef biodiversity survey
campaigns carried out in north Israel representing the first extensive,
quantitative, multiple‐assemblage (including fish, invertebrates and
macroalgae) investigation of Levant reefs that can help define the
ecological status of the south‐eastern portion of this basin. The study
aimed to: (1) assess the impact of fishing on the reefs by comparing
the only functional no‐take reserve (the small Rosh‐Hanikra Achziv
no‐take reserve, with full protection against any kind of fishing for over
a decade) in Israel to nearby areas outside the reserve as well as unpro-
tected reefs located further away (the Haifa Bay area); (2) examine the
percentage cover of the main space occupiers of the reef, focusing
mainly on the presence of brown algae (which are a favourite food of
2RILOV ET AL.
the invasive rabbitfish) and turf (that usually indicates a degraded state,
Connell, Foster, & Airoldi, 2014); and (3) examine the current preva-
lence of alien species on the reefs focusing mainly on fish, algae and
molluscs. Several a priori expectations were postulated: (1) an exten-
sive presence of alien fish in the assemblage, and widespread low
abundance of commercial fish species (except in the small reserve)
because of extensive fishing pressure; (2) low cover of brown‐algae
meadows, and the domination of turf barrens because of rabbitfish
over‐grazing; (3) a domination of invasive molluscs (bivalves and
gastropods) on the reefs because of the scarcity of native molluscs,
and the high rates of invasion of this group in the Mediterranean
(Gofas & Zenetos, 2003; Rilov, 2016). Sampling was conducted in the
shallow waters at several depth strata to also test for potential effects
of depth on the assemblages, and in the Haifa area surveys were
conducted in two seasons (autumn and spring) allowing examination
of the possible effects of temporal variability (period), especially with
regard to brown algae cover, which can be very seasonal.
2|METHODS
2.1 |Study area and sites
Along the coast of Israel, rocky reefs dominate the shallow continental
shelf, mostly in the northern shores. There are two major forms of
subtidal reefs: submerged eolianite (kurkar) ridges (lithification of
ancient dunes that formed on the shoreline during low sea level
periods) that run parallel to the shore, and the submerged portion of
two limestone headlands: Carmel Head in Haifa and Rosh Hanikra near
the Israeli‐Lebanese border. Most of the bedrock on the Israeli coast is
covered by complex biogenic structures made of shells of organisms
and macroalgal crusts. The reefs had no effective protection from
fishing except in one small protected area near the Israeli–Lebanese
border, the Achziv‐Rosh Hanikra nature reserve. In this study, the
sampling focused on reefs in northern Israel in two areas: (1) small
rocky islets (called ‘Achziv‐Rosh Hanikra Islands’) located about 1 km
offshore inside and outside the Achziv‐Rosh Hanikra reserve area;
and (2) the southern parts of Haifa Bay including two kurkar ridges in
the bay itself, and the limestone bedrock of Carmel Head south of
the bay (Figure 1). The Achziv reserve is c. 5 km long and extends from
the shoreline out to 2 km offshore and includes the northern islets.
Only anglers are allowed to fish in the reserve and only from the shore,
but not on the islets where the surveys were done. The Israel Nature
and Parks Authority has a park ranger who has managed the reserve
for almost two decades. The rocky reefs around the Achziv‐Rosh
Hanikra Islands are characterized by a steep and structurally complex
wall below the edge of a vermetid reef that surrounds the islets, down
to about 4–6 m. On the north‐west, west and south‐west faces of the
islets the seascape changes into a shallow, gently‐sloping shelf that is a
few tens of metres wide down to about 12 m. The surface becomes
steeper westward of the shelf, and normally ends with a steep wall
at around 18 m. This wall can be 2–10 m high and at the bottom it
turns into a flat seabed, of either rock or sand. The east face of the
islets is much shallower and flatter. Our surveys focused on the
western faces of the islets.
In Haifa Bay, the depth range of the western and eastern ridges is
16–21 m and 12–18 m, respectively. The Carmel Head has a plateau at
5–10 m depth and its slopes descend to about 20 m (Figure 1). The
shallowest portion of the Israeli coastal reefs (0–5 m), which is mostly
at the base of shoreline‐fringing vermetid reefs (Safriel, 1975), was not
sampled in Haifa in this study, but this section of the reef represents
only a relatively small fraction of the shallow hard‐bottom in Israel. It
should be noted that the Haifa Bay coastline hosts heavy industry
and a big commercial port that has been a source of pollution for
decades; however, pollution levels in most categories have substan-
tially improved over the past two decades all along the Israeli coast
including the Haifa Bay (Herut, 2016). Furthermore, the Bay is very
wide and open (Figure 1) and thus exposed to the main longshore
and cross‐shore currents (ensuring high flushing), and because of this
high water flushing, sampled reefs are presumably relatively little
affected by the distant sources of pollution on the shoreline.
The survey in the small Achziv‐Rosh Hanikra Islands area (Figure 1)
was conducted first, and was relatively limited in scope (i.e. only fish
density and benthic cover assessments, and with lower taxonomic
resolution for some groups; e.g. groupers were not separated into
species). It was conducted over five consecutive days in spring (May)
2011 at two stations inside (the islets of Shahaf and Nachlieli) and
two stations outside (the islets of Akhziv and Shgavion) the marine
reserve, and at each station three depth strata were surveyed: Very
Shallow (the wall at 1–5 m), Shallow (the horizontal shelf at 6–12 m)
and Deep (the wall or slope at 18–24 m). The distance between the
islets inside and outside the reserve is about 1.5 km and the distance
between stations within each area was c. 200 m. Each location/depth
strata included between 4 and 12 fish transects (more in the shallower
depths) and 2–4 benthic cover transects.
The surveys in Haifa Bay–Carmel Head area were conducted
twice, once in the autumn (November 2011) and once in spring
(April–May 2012), at ten hard‐bottom locations (Figure 1). Three loca-
tions were in the bay itself: two on the western ridge (WR) and one on
the eastern ridge (ER); the other seven were in different areas and
depths on top and around the Carmel Head as follows. One in the
centre of the Head's Plateau (PL), and two (one at the top and one at
the bottom of the slope) in each of the three faces of the Head: the
Spartan Reef (SP) in the east, the North Slope (NS), and the Ridge in
the west (RD). The Haifa locations were from two depth strata:
Shallow (S, depth range of 5–12 m) and Deep (D, depth range of
13–20 m). The shallow depth range corresponds to the one used by
Sala et al. (2012) in their Mediterranean‐wide survey (8–12 m). The
ten locations were selected to represent a wide range of potentially
different reef habitats, i.e. different depths, macro‐scale structural
features (ridge, slope, plateau) and areas (the Head vs. the ridges).
In the more extensive Haifa Bay–Carmel Head surveys, each of
the 10 locations included surveys at four main sites serving as
replicates. The sites within each location were separated by several
hundred metres and were chosen haphazardly using a high resolution,
multi‐beam bathymetric map. At each site, the percentage cover of
macroalgae and sessile invertebrates and the density of mobile, highly
patchy or heavily‐fouled invertebrates (focusing mostly on molluscs)
was assessed along transects using counts (for fish and molluscs),
photoquadrats (for percentage cover) or collection of rocks (for
RILOV ET AL.3
epilithic and endo‐cryptolithic biomass). For fish density, more replica-
tion was needed to represent a location because of their high mobility
and patchiness; therefore, 1–3 transects were run at each site, totalling
10–12 replicate transects per location. The distance between fish
transects within a site was 5–20 m. At five sites belonging to three
locations with high structural complexity, pieces of live rock were
sampled to assess the diversity of more cryptic biota (focusing on
molluscs and macroalgae) that is mostly missed by the other sampling
methods (see details below). Transects at each site ran parallel to the
slope to maintain a relatively constant depth; however, owing to the
nature of the reef structure some transects were relatively flat and
some had high structural complexity. All sites were marked by pegs
and float markers designating the main transect (on which all survey
types were done) during the autumn 2011 survey, and a depth profile
was created by measuring bottom depth with a depth gauge (±10 cm)
every metre along the transect tape (see profiles demonstrating the
variety of depths and bottom complexities in Figure S1, Supplementary
material). Data for the two regions were analysed separately.
2.2 |Fish assemblages
Benthic fish assemblages were evaluated by fish counts using scuba
visual surveys (Nunez‐Lara, Arias‐Gonzalez, & Legendre, 2005). The
survey was conducted along 2×30 (60 m
2
) metre belt transects. A diver
swam at a steady speed about 1.5 metre above the bottom holding a
2 m wide pole (to delineate the width of the transect) while deploying
FIGURE 1 Study region and its bathymetry (produced by IOLR from multi‐beam sonar surveys). (a) Map of the Israeli coast with all the known
rocky bottom shown as dark areas. The study regions are shown with a white rectangle. (b) The Levant basin with the study area marked. (c)
The study area of the Achziv‐Rosh Hanikra Islands. The islets are shown in dark shading and the reserve boundary is shown with a broken line. (d)
and (e) are 3D maps of the area viewed from the south and north with the transects marked with bars (cover in tall dark bars and fish in pale short
bars). (f) The Haifa surveys area where the 10 studied locations are marked by ovals, and each location has four main replicate sites marked by
white squares and labels (total 40 sites). (g) Photo showing the complex structure of the reef typical to many areas (the photo was taken on the
North Slope by Hagai Nativ)
4RILOV ET AL.
the transect tape and counting the fish within the transect in front of
him up to 2 m above the bottom (minimum visibility requirement was
5 m). The number of fish in large shoals was estimated to the nearest
10 or 50 individuals (depending on the size of the shoal). The fish
surveys were conducted during daytime (at least 2 hours after dawn
and 2 hours before dusk) and from some distance from the bottom
(about 1 m), therefore the more cryptic and small species (e.g.
blennies, gobies) as well as nocturnal species are underestimated,
especially in high‐complexity areas. Therefore small, cryptic species
were excluded from the analysis. Using scuba gear can also affect fish
behaviour and may deter some species (Lobel, 2005; Watson &
Harvey, 2007); and therefore there may be an under representation
of such species in the data.
2.3 |Macroalgae and sessile invertebrate cover
To estimate the percentage cover of macroalgae and sessile (mostly
encrusting) invertebrates, 15 random 50×50 cm photoquadrats
(using a random number generator) were taken along the marked
30 m transect at each site (Giakoumi, Cebrian, Kokkoris,
Ballesteros, & Sala, 2012). A Nikon D90 camera with a fisheye lens
was used to enable taking a photo of a 50×50 cm area c.30cm
from the substrate, resulting in high resolution photos. The photos
were straightened with Adobe Photoshop, the percentage cover of
different taxa (species when possible and higher taxonomic levels
or morpho‐functional groups when species could not be identified)
was assessed with the CPCe software (Kohler & Gill, 2006) using
uniform point contact (UPC, 100 points per photo), and an average
cover per transect was calculated. Morpho‐functional groups were
used to classify macroalgae or sponges that could not be identified
in the photos, but their form is obvious and can be used
to distinguish between different cover types (Canning‐Clode,
Maloney, McMahon, & Wahl, 2010). This includes, for example,
categories such as: rhodophyta‐foliose, rhodophyta‐filamentous,
sponge–red encrusting (there are several encrusting red sponges
that cannot be separated visually, only by looking under the micro-
scope), etc.
2.4 |Molluscan assemblage
To estimate the density of mollusc species that are either mobile, very
patchy or heavily fouled (e.g. the rock‐cemented and heavily fouled
bivalves Spondylus spinosus, and Chama pacifica), the species were
counted along a 30×1 m swath of the main transect. The transect
was divided into three 10 m sections that were counted separately
and then analysed jointly. Because of time limitations when working
underwater (usually 45 minutes in this study's depths) it was not
possible to effectively count all individuals of highly abundant species
(some snails and bivalves). Therefore, subsamples of individuals of
abundant species were counted along each 10 m section until 30
individuals of that species were reached. At that point, the diver wrote
down the distance surveyed in that 10 m section along the transect
tape. This distance was used later to extrapolate the number of individ-
uals of that particular species for the entire 10‐m section. This
unavoidable approach (known as the variable area sampling method
(Seber, 1982)) may result in some overestimation in cases where the
animals are highly patchy. The relative numbers of the two invasive
oysters, S. spinosus and C. pacifica should be taken with caution
because it is difficult to distinguish between them without removal
owing to their plastic morphology, and high shell erosion and biofoul-
ing. Another pair of non‐indigenous oysters, Alectryonella crenulifers
(now accepted as Dendostrea sandvichensis) and Alectryonella plicatula,
which were reported on the Israeli coast in the last decade (the latter
only during this study but probably arrived in the area before) (Mienis,
Zaslow, & Rittner, 2012; Sharon, Benayahu, & Mienis, 2005), were
highly cryptic (only the opening of the valves is usually visible), and
could not be distinguished in the field and thus were united in the
analysis under Alectryonella spp.. As large echinoderms (sea cucumbers
and urchins) are ecologically important and are also very patchy and
mobile, their density was also recorded in the survey.
2.5 |Live rock biota
In the Carmel Head area, the aim was to assess the smaller‐scale and
cryptic diversity in structurally‐complex areas –i.e. cryptic and
encrusting macrobiota in shaded or sheltered areas or those that live
within rocks (boring animals inside the highly porous, biogenic rock
characterizing much of the Israeli rocky substrate) that are missed
using photoquadrats or swath surveys. Here the focus, as in the other
methods, was on macroalgae and molluscs, but other taxa were also
documented, albeit with less taxonomic detail. For that, ‘live rocks’
(mostly biogenic with high cover of biota) were sampled at five
complex sites and subsequently processed in the lab. Five pieces of
rock were broken‐off the substrate with a hammer at two sites at
Spartan Reef, two at West Ridge, and one on the North Slope. The
rocks averaged 1.032 ± 0.49 (SD) litre in volume, 475 ± 815 cm
2
in
surface area and 1.8 ± 0.20 kilos in weight. The rocks were placed in
buckets full of seawater on the boat and were brought live to the lab
and then kept in flow‐through seawater aquariums until processing
(normally, several days later).
Processing of rocks involved several steps: (1) rock weight and
volume calculations; (2) taking photos of 4‐6 faces of the rocks from
which area cover and biomass calculations of the different organisms
was made (cover was assessed with CPCe by marking the area of the
different organisms); (3) removal of individual organisms or parts of
organisms for identification and biomass calculation for both the
epibionts and infauna (mostly worms and bivalves) by carefully break-
ing the rocks to pieces with a chisel and hammer; (4) weighing of
samples for biomass analysis (for encrusting animals and macroalgae,
a piece with a known area was taken and the biomass was calculated
for the total organism cover by extrapolation using the rock photos);
(5) consultation on identification with local taxonomists (if an organism
could not be taxonomically identified to the species level locally, the
identification was done to the highest level possible (class, order,
family or genus)); and (6) calculating percentage cover and biomass
per rock surface of all observed taxa. The only unifying measure for
the analysis of the overall community composition of these different
taxonomic forms was biomass (wet weight), and thus the statistical
analysis is based on biomass per rock surface area.
RILOV ET AL.5
2.6 |Data analysis
To investigate the variability in community assemblage among loca-
tions (including protection level) and between seasons (main effect:
period) and depth strata, non‐parametric, permutation‐based, multi-
variate analysis was applied on the fish and rock cover data using
the PRIMER 6 software package (Anderson, Gorley, & Clarke, 2008).
The data were log (x + 1) transformed to reduce the effect of the
most abundant taxa. Bray–Curtis dissimilarity was calculated, and
non‐metric multidimensional scaling (nMDS) ordinations were pro-
duced to identify clustering of samples. Permutational multivariate
analysis of variance (PERMANOVA) was performed to test for main
effects and interactions, and SIMPER analysis was done in some
cases to test which taxa were mostly driving the variability. The effect
of depth on community structure was tested only in Achziv‐Rosh
Hanikra and on the three faces of the Carmel Head, where the two
depth strata were sampled in close proximity (to avoid potential
spatially‐confounding effects). T‐test was used to compare means
of key fish species between protection levels (in Achziv) and depths
(in Haifa).
3|RESULTS
3.1 |Fish assemblage
3.1.1 |Achziv‐Rosh Hanikra area
In total, 23 taxa (most identified to species and a few to family level)
were observed during spring 2011, of which six were alien species.
Overall, the average fish abundance inside the reserve was 40% higher
than outside (125 vs. 78, t‐test, n = 76, P= 0.009). Commercial native
species (the sparids Diplodus sargus,Diplodus vulgaris and Diplodus
cervinus,Sarpa salpa,Oblada melanura and groupers, which mostly
included Mycteroperca rubra and Epinephelus marginatus) were much
more abundant (by between four‐fold and up to two orders of magni-
tude) at the stations inside the reserve compared with outside the
reserve, mostly around the walls surrounding the islets in the very
shallow waters (Figure 2). For example, D. sargus density on the
shallow wall inside the reserve was 13.8 ± 6.0 (average ± SE) fish per
60 m
2
compared to 1.1 ± 0.6 at this depth outside the reserve (t‐test,
n = 35, P= 0.09). The sparids almost always moved inside the reserve
in schools of different sizes and were present in large numbers in
FIGURE 2 Average density (per 60 m
2
transect) and frequency of occurrence of fish, percentage cover and frequency of occurrence of macroalgae
and sessile invertebrates on the Achziv‐Rosh Hanikra reefs during spring 2012. Only the 14 most abundant cover categories or taxa are shown.
Alien species names are marked with an asterisk and commercial fish species are marked with a plus sign. Bar shadings indicate depth strata (in
metres)
6RILOV ET AL.
several transects and absent altogether in other transects resulting in
marginally non‐significant univariate analysis when testing the effect
of protection on single species, even though means were very differ-
ent. The two invasive siganids interestingly showed contrasting trends
in the shallow waters: Siganus luridus was 10 times more abundant
inside the reserve than outside (68.1 ± 17.9 and 7.1 ± 3.0, respectively,
t‐test, n = 35, P= 0.008), whereas S. rivulatus was twice as abundant
outside the reserve (15.8 ± 6.3 and 33.9 ± 11.6, t‐test, n = 35,
P= 0.09). Consequently, the total density of herbivorous siganids
was higher inside the reserve and they also often moved in large mixed
schools. Overall, alien species represented between 23 and 44% of
the total fish abundance in the area, and did not differ between loca-
tions inside and outside the reserve. By contrast, commercial species
represented 21.5% of the total fish abundance in the shallow water
assemblage inside the reserve compared with only 7.4% outside the
reserve.
The PERMANOVA results show that Depth and Depth × Protec-
tion level (locations inside or outside the reserve) had significant
influence on the fish assemblage (Table 1). The interaction indicates
that assemblages were significantly different between areas inside
and outside the reserve only at certain depths: on the walls found in
both the very shallow and the deep depth strata but not on the shal-
low shelf (t = 2.4, P= 0.0002, t = 1.5, P= 0.04, and t = 0.8, P= 0.65,
respectively; PERMANOVA pairwise analysis), as can be seen in the
MDS plot (Figure 2). However, the shallow shelf, with its much lower
structural complexity, had very low densities of commercial species in
both areas.
3.1.2 |Haifa Bay area
In total, 35 taxa (most identified to species and a few to family level)
were observed during the two survey periods, of which nine were alien
species (Table S1 in the Supplementary Material). Native species
targeted by fishermen occurred in only a fraction of the transects in
both seasons (usually 10% or lower) and at very low average densities
(Figure 3). Alien species constituted 36% and 26% on average of all
individuals counted, during the autumn and spring surveys respec-
tively. On average, the most numerically abundant fish was the
damselfish Chromis chromis (mean of 23 and 8 individuals per transect,
in the autumn and spring respectively, Figure 3). It occurred in 61% and
50% of the transects, in the autumn and spring, respectively. This small
planktivorous species usually occurred in large shoals sometimes up to
several metres above the bottom. The two alien rabbitfishes were
next in average density and occurred in 35–49% of the transects
(Figure 3). Native herbivores, Sarpa salpa and Sparisoma cretense, were
completely absent during these surveys (but these species were seen,
albeit rarely, in later surveys in north Israel, Rilov, unpublished data).
The species that occurred in most transects are two native wrasses,
TABLE 1 PERMANOVA results on the effect of Protection and Depth on the reef fish assemblage and the reef macro‐benthos cover in the spring
2011 Achziv‐Rosh Hanikra survey
Fish
Source df SS MS Pseudo‐F P(perm) Unique permutations
Protection level 1 2412.9 2412.9 2.0462 0.0706 9942
Depth 2 26278 13139 11.142 0.0001 9931
PrxDe 2 6742.2 3371.1 2.8587 0.0017 9931
Residuals 71 8.37E + 04 1179.2
Total 76 1.22E + 05
Estimates of components of variation
Source Estimate Sq. root
Protection level 39.07 6.2503
Depth 510.2 22.587
PrxDe 187 13.675
Residuals 1179 34.34
Macro‐benthos cover
Source df SS MS Pseudo‐F P(perm) Unique permutations
Protection level 1 4093.4 4093.4 3.8177 0.0014 9946
Depth 2 8259.4 4129.7 3.8516 0.0002 9932
PrxDe 2 2653.2 1326.6 1.2373 0.2544 9920
Residuals 23 24661 1072.2
Total 28 40704
Estimates of components of variation
Source Estimate Sq. root
Protection 222.4 14.913
Depth 331.7 18.213
PrxDe 55.2 7.4298
Residuals 1072 32.745
RILOV ET AL.7
Thalassoma pavo and Coris julis. Also abundant was another alien, the
more cryptic, mostly nocturnal, squirrelfish, Sargocentron rubrum
(Figure 3), which was present in small groups mostly inside or nearby
crevices and small caves. A recently‐arrived nocturnal alien species,
the cardinalfish, Cheilodipterus novemstriatus (Goren, Lipsky, Brokovich,
& Abelson, 2010), occurred only in one transect on the East Ridge
during the autumn survey but in high numbers (55 individuals). The
rest of the species occurred at very low abundances (<0.5 individuals
on average per transect) including all commercial species except one,
Diplodus vulgaris, which did occur in one transect in relatively high
numbers (34 individuals in the autumn) and had an average of 0.78
individuals per transect in that season. All apex predators were
extremely rare. Groupers were absent in most transects (occurring in
44 of the 235 transects, 18%), but on three occasions 6–8 individuals
were seen in the deeper parts of Carmel Head (RD‐D and SP1 sites,
Figure 3). Notable also was the occurrence of 24 individuals of the
round stingray, Taeniura grabata, in one transect during spring; proba-
bly a reproduction aggregation.
The PERMANOVA analysis indicates that both Location and
Period and their interaction had strong and significant effects on
community structure (Table 2). Analysis of the subset of the data (the
three faces of the Carmel Head) indicates that both reef Face and
Depth Strata in the two seasons, as well as their three‐way interaction,
significantly contributed to the fish assemblage variability, but Depth
had the strongest contribution (Table 3). The four species that most
contributed to the difference between depths (as indicated by the
SIMPER analysis on the log‐transformed data, Table 4) were Chromis
chromis (t‐test, n = 233, P= 0.0004), Siganus luridus (t‐test, n = 233,
P< 0.0001) and Thalassoma pavo (t‐test, n = 233, P< 0.0001),which
were more abundant in the shallower depth (5–12 m), and Coris julis
(t‐test, n = 233, P< 0.0001), which was more abundant in deeper
waters (13–20 m).
FIGURE 3 Average density (per 60 m
2
transect) and frequency of occurrence of fish, percentage cover and frequency of occurrence of macroalgae
and sessile invertebrates on the reefs and average density (per 30 m
2
transect) and frequency of occurrence of molluscs during autumn 2011 and
spring 2012. Only the 15 most abundant species or taxa are shown. Alien species are in black bars, turf is shown with a hashed bar. Commercial fish
species are indicated with an asterisk
8RILOV ET AL.
TABLE 3 PERMANOVA results on the effect of ‘Face’, Depth strata and Period on the reef fish assemblage
Source df SS MS Pseudo‐F P(perm) Unique permutations
Face 2 7796.3 3898.2 3.0862 0.0027 9930
Depth 1 8881.7 8881.7 7.0316 0.0002 9945
Period 1 5546.3 5546.3 4.3909 0.0028 9945
FcxDe 2 7406.9 3703.4 2.932 0.0026 9931
FcxPe 2 4337.8 2168.9 1.7171 0.0852 9936
DexPe 1 2865.5 2865.5 2.2686 0.0537 9957
FcxDexPe 2 8014.4 4007.2 3.1725 0.0024 9931
Res 115 1.45E + 05 1263.1
Total 126 1.91E + 05
Estimates of components of variation
Source Estimate Sq. root
Face 62.966 7.9351
Depth 121.73 11.033
Period 68.436 8.2726
FcxDe 116.63 10.799
FcxPe 43.289 6.5795
DexPe 51.206 7.1558
FcxDexPe 262.29 16.195
Residuals 1263.1 35.54
TABLE 2 PERMANOVA results on the effect of Location and Period on the reef fish assemblage
Source df SS MS Pseudo‐F P(perm) Unique permutations
Location 9 65225 7247.2 6.2338 0.0001 9891
Period 1 7663.4 7663.4 6.5917 0.0001 9948
LoxPe 9 25987 2887.4 2.4836 0.0001 9875
Residuals 192 2.23E + 05 1162.6
Total 211 3.23E + 05
Estimates of components of variation
Source Estimate Sq. root
Location 289.95 17.028
Season 62.288 7.8923
LoxSe 164.39 12.821
Residuals 1162.6 34.097
TABLE 4 SIMPER results showing the fish species contributing most to the dissimilarity between depth strata in the three tested ‘Faces’of the
Carmel Head in Haifa Bay (the two seasons are combined). The first two columns are log transformed averages of the species at the different
depths. The average dissimilarity between depths was 53.5%
Depth strata Deep Shallow
Species Avg Abundance Avg Abundance Avg Dissimilarity Diss/SD Contribution % Cumulative %
Chromis chromis 1.65 1.80 10.05 1.21 18.79 18.79
Siganus luridus 0.78 1.63 8.08 1.18 15.11 33.91
Coris julis 2.13 1.49 6.99 1.09 13.06 46.97
Thalassoma pavo 1.77 2.48 6.84 0.99 12.79 59.76
Sargocentron rubrum 0.69 0.65 4.79 0.88 8.96 68.73
Siganus rivulatus 0.54 0.57 4.36 0.79 8.15 76.88
Serranus cabrilla 0.36 0.15 2.41 0.77 4.50 81.38
Serranus scriba 0.26 0.26 2.04 0.81 3.81 85.19
Epinephelus marginatus 0.23 0.13 1.51 0.57 2.83 88.02
RILOV ET AL.9
3.2 |Macroalgae and sessile invertebrate cover
3.2.1 |Achziv‐Rosh Hanikra area
The top space occupiers on reefs in both areas were turf algae and the
red calcifying algae Amphiroa spp. In the literature, the term ‘turf’refers
to several forms of low‐lying benthic cover (Connell et al., 2014), and
this study broadly refers, as in most cases, to a matrix of low‐lying algal
mats (usually <2 cm in height), which is usually a mixture of filamentous
(e.g. Polysiphonia) or calcareous articulated or coarsely branched
(mostly grazed corallines) taxa, with much sediment trapped within
the matrix. It is presumably similar to what was previously described
by Sala et al. (2011) for Mediterranean reefs, as well as by Vergés,
Tomas, et al. (2014) who described it also as ‘epilithic algal matrix’. Turf
occupied between 28 and 71% of the reef. The shallow reef zones
inside the reserve had higher cover of fleshy brown algae such as
native species Taonia atomaria (16.4% on average), Dictyota spp.
(7.3%) and Sargassum spp. (2.5%) as well as the invasive Stypopodium
schimperi (10.4%); all of which were almost absent on the reefs outside
the reserve (no higher than 0.6%) during spring 2011. The two alien
Indo‐Pacific bivalves that could not be distinguished in the photo-
graphs, Spondylus spinosus and Chama pacifica had the highest cover
(21.5%) in the deepest zone outside the reserve. The PERMANOVA
analysis reveals that both Protection level and Depth had a strong
effect on the structure of the macrobenthos cover (Table 1). A cluster-
ing by Protection level and Depth is evident in the MDS plot (Figure 2),
and, although the PERMAOVA could not detect an interaction
between Protection and Depth, the greatest separation between tran-
sects within and outside the reserve appears to be in the shallower
depths where brown macroalgae were more abundant inside the
reserve.
3.2.2 |Haifa Bay area
As in Achziv, the top space occupiers on the reef were turf algae and
encrusting corallines that together occupied, on average, 45–47% of
the rock surface (Figure 3). At many sites, turf covered nearly 90% of
the reef surface (Figure 5j). Next were the two alien bivalves, S.
spinosus and C. pacifica, with 3.7 and 0.73% percentage cover, in the
autumn and spring, respectively. They also occurred in almost all tran-
sects (Figure 3). These cover values are probably an under‐estimation
because these bivalves are frequently heavily fouled, and thus are hard
to detect in the photoquadrats, mostly so in the spring when more
algae grow on the reef. Additional rock cover included other forms of
red but also green and brown algae, many sponges, bivalves, ascidians,
hydrozoans and bryozoans, all with relatively low coverage. Notably,
the lowest cover among macrophytes was that of brown algae
(Phaeophyta) that were not present at all in the autumn survey, and
were present, but with very low average cover, in the spring. These
species include Dictyota spp. (Figure 5b) Dictyopteris polypodioides
(Figure 5c), Halopteris scoparia (Figure 5d),as well as the Fucales,
Sargassum spp. and Cystoseira spp. In several of the shallower sites
on the Carmel Head (mostly on the shallow Ridge) there was high
cover of some of these species (e.g. H. scoparia reached 50% in one
site) during spring, but when averaged over all sites their abundance
was very low (<3%, all species combined). Some alien macroalgae had
high coverage (20–60%) but only at very few sites. These include
Codium parvulum,Galaxaura rugosa (both, mostly found in the Deep
Ridge) and S. schimperi (mostly in the spring among the native brown
algae in the Shallow Ridge), as well as a suspected alien, an unidentified
Laurencia species not described in the region in the past. Other
frequently‐encountered aliens were the large, branching, hydrozoan
Macrorhynchia philippina, and the relatively cryptic bivalve Malleus
regula. When applying the macroalgal (functional) grouping used by
Sala et al. (2012) to the data, the average rock area covered by turf
in autumn was 43%, 5.1% was covered by encrusting forms, 3% by
alien turf, 2.1% by erect forms, 1.9% by erect aliens, and 0.002% by
native canopy algae. In the spring, turf covered 43%; 6.4% was covered
by encrusting algae, 4.2% erect algae, 1.8% erect alien algae, and 0.5%
canopy algal forms.
The PERMANOVA analysis reveals that both Location and Period
had a strong effect on the structure of the macrobenthos cover
(Table 5). Their interaction was also significant, and is driven mostly
by the fact that the spring high cover of fleshy brown algae was seen
only in the shallow sites on the Carmel Head Ridge (sites circled by
an oval in Figure 4a).
3.3 |Molluscan assemblage
Aliens completely dominated the epi‐benthic molluscan assemblage on
the Haifa Bay and Carmel Head reefs (Figure 3). Among the bivalves,
99.9 and 100% (in the autumn and spring, respectively) of all individ-
uals counted in the transects were of Indo‐Pacific origin, and for
gastropods 95.5 and 97.5% were aliens. Among the bivalves, the most
abundant species were the relatively cryptic hammer oyster, M. regula,
and the oysters, S. spinosus and C. pacifica. In some sites, the two
oysters formed 3‐dimentional structures (made of both live and dead
individuals) on top of the reef owing to their high density there
(Figure 5i). Among the gastropods, the cerithid Cerithium scabridum
and the strombus Conomurex persicus were the most abundant. These
snails mostly occurred on rock areas covered by turf and sediment and
were highly patchy in their distribution. It should be noted that within
thick macroalgae patches, high density of the native snail, Columbella
rustica, can sometimes be found (in algal samples taken to the lab,
unpublished data), but this is mostly missed in regular visual surveys.
It should also be noted that the only large echinoderm encountered
in the transects was the invasive sea cucumber Synaptula reciprocans,
with relative frequency of 40% in the autumn, and up to around 300
individuals in one of the sites. No sea urchins were observed in the tran-
sects but a few were seen outside them (less than 15 in near 100 dives).
3.4 |Live rock biota
Taxa abundance (Figure 6, showing only the 10 highest ranked) in
the samples is ranked based on two categories, average biomass per
rock per season and frequency of occurrence on the live rock (out of
25 rock samples per period). During the autumn sampling, the reef‐
building oyster Spondylus spinosus was the third most contributing taxa
in terms of average biomass because it is a very large species but it
occurred on only 17% of the rocks and only at some sites. It was
completely missed during the spring sampling (probably because of
10 RILOV ET AL.
its very patchy distribution) with zero occurrence, although it was
recorded in the molluscan count and in the photoquadrats during this
sampling period (see above). In contrast, the other, smaller, alien oyster
C. pacifica was only 7
th
in average biomass in the autumn but occurred
on almost half the rocks (frequency of occurrence = 47%). In the spring,
it was second in biomass with Frequency = 71%. Another alien bivalve,
M. regula occurred in 46 and 42% of the rocks in the autumn and the
spring respectively. Interestingly, although native molluscs were
incredibly rare over the reef surface (see above), cryptic native bivalves
were abundant within the rocks. The native rock‐boring bivalve,
Lithophaga lithophaga was found in most rocks in both seasons and
also contributed much to the biomass (Figure 6). Likewise, the bivalve
Striarca lactea was very abundant in both seasons, although it contrib-
uted little to the biomass owing to its small size.
Turf occurred on all rocks in both seasons with a biomass that
exceeded the second most abundant taxa by more than an order of
magnitude. Two red crustose algae labelled CCA1 (hard crust) and
CCA2 (softer crust, mostly from the genus Peyssonnelia) were also very
abundant during both seasons. One identified corticated red alga,
Chondracanthus acicularis, was abundant in autumn but less abundant
in the spring. The invasive alga Galaxaura rugosa occurred on 29%
and 58% of the rocks in the autumn and spring, respectively, and its
mean biomass increased 30‐fold.
4|DISCUSSION
There is no information or documentation of what pristine nearshore
reefs looked like in the Levant. However, even without prior quantita-
tive data for most biotic groups, the results of this study strongly
suggest that the reefs in the region of the south‐eastern Mediterra-
nean studied here have been dramatically transformed in the past
century, and are now overfished, dominated by aliens and turf, and
sustain low levels of habitat‐forming (canopy) algae, probably due to
overgrazing by invasive herbivorous fish.
4.1 |Fish assemblages: Overfishing and alien
prevalence
The surveys showed that species that are targeted by fishermen
(large benthic predatory and omnivorous species, as well as the
native herbivore S. salpa),represent a very small part of the fish
assemblage at all sites other than the small Achziv reserve. Because
it appears that most of these large native reef fish can still thrive in
this fast‐warming region when protected (as evident by their high
proportion, 22%, inside the reserve), it seems that climate change
has so far played little if any part in reducing their populations. This
does not mean that with additional warming they will not be
TABLE 5 PERMANOVA results on the effect of Location and Period on the reef macro‐benthos cover assemblage
Source df SS MS Pseudo‐F P(perm) Unique permutations
Location 9 25877 2875.2 2.5869 10.0001 9819
Period 1 8171 8171 7.3517 0.0001 9929
LoxPe 9 14331 1592.4 1.4327 0.0162 9838
Residuals 62 68910 1111.5
Total 81 1.17E + 05
Estimates of components of variation
Source Estimate Sq. root
Location 219.79 14.825
Period 177.96 13.34
LoxPe 119.86 10.948
Residuals 1111.5 33.338
FIGURE 4 nMDS ordination of the centroids
of locations per period of the fish assemblages
demonstrating the clear difference between
the shallower and deeper parts of the reef
RILOV ET AL.11
affected. The paucity of target species outside the reserve both in
Achziv and the Haifa area is therefore a clear indication that the rest
of the reefs are highly overfished. The driver is most probably the
unregulated, unreported, artisanal and mostly recreational fishing that
has substantially increased in the last decade or two (Edelist,
Scheinin, et al., 2013). The low predator abundance on the Israeli
reefs is similar to what was found at all unprotected sites in the rest
of the Mediterranean (Sala et al., 2012), indicating that reefs in the
entire Mediterranean are under heavy fishing pressure including the
south‐eastern Levant.
Aliens represented a considerable portion of the fish assemblage
(average 23–44% of fish density per site, excluding cryptic and noc-
turnal species). The most dominant ones were the two Red Sea
rabbitfish, S. rivulatus and S. luridus that were first documented on
the Levant reefs in 1924 and 1955, respectively (Ben‐Tuvia, 1964;
Steinitz, 1927). Shallow reef fish surveys conducted between the
FIGURE 5 nMDS ordination of the rock cover assemblages in 40 sites and two periods (a) demonstrating mostly how one location (the shallow
ridge in the Carmel Head) stands out during spring when the reef in that area was covered by a meadow of brown algae shown in bubble
graphs indicating percentage cover of different species (b‐d). The cover by alien algae (e‐h) and bivalves (i) –marked with * (** = suspected alien) –
was very patchy as well, but turf (j) was highly dominant in almost all locations
12 RILOV ET AL.
mid‐1970s and early 1980s in an area 25 km south of Haifa (Diamant,
Bentuvia, Baranes, & Golani, 1986), revealed a relatively low propor-
tion of alien presence out of the total assemblage (7.4%) including
low numbers of rabbitfish. But much higher proportions of aliens were
reported at a site in Haifa in the 1990s (Goren & Galil, 2001), and in
the early 2000s at several sites along the Lebanese coast (Harmelin‐
Vivien et al., 2005), with especially high presence of rabbitfish. It
appears that the average proportion of aliens in fish counts has
stabilized around 30–40% in the past two decades, although they
can represent >50% of the total fish biomass (Goren & Galil, 2001).
Although the total density of fish was higher in the Achziv reserve
compared with unprotected areas, aliens maintained a similar propor-
tion within the fish assemblage inside the reserve as that found in
fished areas. This was mostly because S. luridus was more abundant
inside than outside the reserve (in shallow waters). Theoretically, one
would expect that the higher abundance of predators in the reserve
would help control the invasive herbivore population and thus release
macroalgae (brown algae meadows) from grazing pressure through a
trophic cascade (Guidetti, 2005; Pinnegar et al., 2000). However, this
mechanism evidently does not always apply in marine reserves
(Cardona, Sales, & Lopez, 2007) as evident in our study. No effect of
MPAs on non‐indigenous species abundances was also observed
throughout the northern parts of the Mediterranean (Guidetti et al.,
2014). One possibility is that the piscivores (mostly groupers) in the
reserve were not yet abundant enough or large enough to control
the rabbitfish population, as Siganus was shown to be the most
abundant item (68% of the biomass) in the diet of the grouper M. rubra
(Aronov & Goren, 2008). The higher cover of brown algae meadows
measured inside the reserve compared with outside could be a result
of natural variability among sites, or a non‐trophic effect of the preda-
tors on the siganids through behavioural changes (e.g. less active, more
timid, in the presence of many predators). This aspect should be tested
experimentally in future studies.
4.2 |Rock cover: The domination of turf, paucity of
brown algae meadows and rise of aliens
The data show how reefs in northern Israel today are dominated by
turf barrens (turf domination of rock cover was seen at both transect
scale as well as the smaller live rock scale). This is a clear indication
of the poor ecological state of the system, as turf normally character-
izes benthic ecosystems under stress (Connell et al., 2014). In many
regions, shifts to turf barrens on coral or rocky reefs are driven by
eutrophication, global change, high sediment load and/or overgrazing
(Falkenberg, Connell, Coffee, Ghedini, & Russell, 2015; Jouffray et al.,
2015); the latter being probably the main driver in the study system.
Every other space occupier on the reefs in the study region was found
to be highly patchy, including most canopy and erect macroalgae, both
native and non‐native (Figure 5). Erect (mostly Dictyota, Dictyopteris
and Halopteris spp.) and canopy (the Fucales, Cystoseira and Sargassum
spp.) brown algae were found only in very few transects in the shallow
waters of the Carmel Head, as well as inside the Achziv‐Rosh Hanikra
marine reserve, and fucales only in the spring time (at the end of the
contemporary main growing season, winter; Rilov, unpublished data).
Sites with high brown algae cover were also characterized by no or
FIGURE 6 Live rock analysis. Average biomass and frequency of
occurrence of taxa or functional groups during autumn 2011 and
spring 2012 at five sites (n = 5 rocks per site). Only the 10 most
abundant species or taxa are shown. Alien species are in black bars,
turf is in a grey bar
RILOV ET AL.13
very low numbers of siganids and a very simple seascape. Observations
and surveys in other areas along the Israeli coast (Rilov, unpublished
data) also indicate that small meadows of brown algae are mostly
restricted to relatively flat seascapes with scarce refugia (holes, crev-
ices, caves) and with no siganids. It is possible that in low‐complexity
seascapes, relatively large fish like the herbivorous rabbitfish are
absent because of rarity of refuge and thus have high predation risk.
We suggest that rabbitfish avoid open reef areas with little refugia
because they are wary of travelling too far from nearby refuge due
to fear of transient predators (such as jacks that can still be seen occa-
sionally, but are mostly missed in our surveys), and thus brown macro-
phyte meadows stay intact in such seascapes. A similar seascape effect
was shown in coral reefs, where macrophyte meadows have distinctive
halos (clearings) around coral patches, halos that are created by herbiv-
orous fish that find refugia in the coral patches and roam and graze
only at short distances from them (Downie, Babcock, Thomson, &
Vanderklift, 2013; Madin, Madin, & Booth, 2011).
Interestingly, even within existing meadows, canopy‐forming algae
(the Fucales Cystoseira spp. and Sargassum spp.) that characterize
healthy shallow Mediterranean reefs (Guidetti & Dulcic, 2007; Guidetti
& Sala, 2007) were rather rare. In a Haifa Bay study that included
monthly benthic dredge or a bottom grab sampling between April
1955 and October 1956, 62 samples were taken from rocky areas at
depths ranging from 18–50 m, and both Cystoseira and Sargassum
occurred in 25% of these samples, indicating that they were indeed
once much more abundant in the region (Edelstein, 1960). It should
be noted though that a meadow of Fucales has been annually observed
in very shallow waters (1–4 m) in Haifa at the Shikmona beach
between January and May (Rilov, unpublished data) and there may
be other meadows found along the coast. According to Edelstein
(1960), in the 1950s the canopy‐forming Cystoseira was in full growth
in the summer (August) months, but today, it starts deteriorating in late
May, and between late June and January it is only found in its dormant,
branchless, form (Rilov, unpublished data), suggesting that its growth
season has contracted considerably, possibly due to ocean warming.
Indeed, recent experimental work using novel mesocosm methods
(Wahl et al., 2015) showed that when temperatures are cooled
by two degrees (reconstructing temperatures in the 1980s) shallow
subtidal Cystoseira from the Israeli coast deteriorated much more
slowly in late spring and early summer than in ambient waters (Guy
Haim, 2017). Clearly, at the coastal scale (ecosystem level), it is reason-
able to assume that overall benthic primary productivity (and probably
the entire food web) has reduced considerably in the south‐east
Mediterranean owing to the decimation of most brown macroalgae
following the siganid invasion, and more recently also owing to ocean
warming. Recent in situ metabolic measurements of different com-
munity types on the shallow Israeli reefs indeed showed that net
primary production is almost an order of magnitude lower in turf plots
compared with plots dominated by brown algae canopy (Peleg, 2016).
The distribution of alien macroalgae was also extremely patchy
(Figure 4). This included mostly the three Indo‐Pacific species,
Galaxaura rugosa (first discovered in the Mediterranean in Syria in
1999, Bitar et al., 2017; first discovered in the Mediterranean in Israel
in 2003, Hoffman, Israel, Lipkin, Dubinsky, & Iluz, 2008), Codium
parvulum (found in Lebanon in 2008, where it is reported to be invasive
along the entire coast, Bitar et al. (2017); but first discovered in the
Mediterranean in Israel in 2004, Israel et al. (2010)), and Stypopodium
schimperi (already found in Syria in 1979 and in Lebanon in 1991, where
it is now highly abundant all along the coast, Bitar et al. (2017); first
reported in Israel in 1996, Lundberg (1996)). Large patches of G. rugosa
(found either as monoculture or mixed with native macroalgae) were
mainly present at around 10 m depth, dense carpets of C. parvulum
were mainly present on deeper slopes (15–20 m deep), and S. schimperi
was mainly found in the shallower depths among erect native
macroalgae, mostly in the Achziv area. During winter and spring, large
drifts of these species can often be found along northern shores follow-
ing storms (Hoffman et al., 2008; Israel et al., 2010), indicating the large
biomass that these species can reach. The first two species were
reported from the Israeli shore only a decade ago, suggesting the fast
growth of their populations in the region. Why the distribution of these
invaders is so patchy is unknown. It is most probably not related to
competition with native macroalgae, as most of the reef is now covered
by turf and presumably open for invasion (Arenas, Sánchez, Hawkins, &
Jenkins, 2006). We suggest that it is related to the mode of reproduc-
tion and spread, as the dispersal distance of propagules of most
macroalgae is relatively restricted (Norton, 1992). New patches pre-
sumably establish as a result of propagules or spores released from
drifting algae after storms, and thus, with time, it is expected that these
species will spread westward in the Mediterranean (none of them is
currently known to have established large populations outside of the
studied region). Conversely, why the thermophilic, highly invasive,
Caulerpa racemosa and Caulerpa taxifolia, that have ‘taken‐over’many
reefs in much of the Mediterranean including Cyprus (Klein & Verlaque,
2008), have not established on the Israeli reefs (although they have
been reported in Lebanon, Bitar et al. (2017)) is an enigma.
4.3 |Molluscan assemblages: An overwhelming alien
domination
Aliens overwhelmingly dominate the epi‐benthic molluscan fauna
(as well as that of large echinoderms, namely the IndoPacific sea
cucumber, S. reciprocans). The only habitat where native bivalves
(the boring species, Lithophaga lithophaga and Striarca lactea) are still
abundant is inside the rocks where they solely dominate this cryptic
niche. The two large alien, rock‐cemented, Indo‐Pacific bivalves,
Spondylus spinosus (reported as well established two decades ago,
Mienis, Galili, and Rapoport (1993); and again in a 2010 study,
Shabtay, Tikochinski, Benayahu, and Rilov (2014) and Chama pacifica
(first reported in Egypt in 1905 and later in Israel in 1998 and spread-
ing as far as Greece in the past decade, Crocetta and Russo (2013)), not
only dominate numerically and in biomass (as indicated from the live
rock analysis), in some areas along the coast they can even form
multi‐layered structures that considerably add to the complexity of
the reefs and probably alter their function. Similarly, the Indo‐Pacific
detritivore snail, Cerithium scabridum (one of the first aliens to cross
the Suez Canal and located in Egypt in 1883 and since then well
established in the Levant coast, Zenetos, Ovalis, and Kalogirou
(2009)) and the Persian Gulf herbivorous snail Conomurex persicus (first
sighted in Turkey in 1978, and then in rapid succession off Israel,
Rhodes, Cyprus and Lebanon; Nicolay and Manoja (1983)), now
14 RILOV ET AL.
dominate the reef gastropod assemblage. A recent study has shown
that in the Mediterranean there are many more established Red Sea
bivalves on rocky substrates than in soft bottoms, and the authors
speculated that it might be because of lower biotic resistance in hard
bottoms, i.e. fewer natives (Nawrot, Chattopadhyay, & Zuschin,
2015). However, many native reef molluscs, including species from
the same genus as the invaders, were once abundant on the Israeli
reefs, as described in the taxonomic literature (Barash & Danin,
1982). Why almost all of them have disappeared is difficult to answer,
but the reasons could include change in the physical conditions, mainly
extensive warming, as was shown for urchins (Yeruham et al., 2015)
and suggested also for molluscs (Rilov, 2016), competition for
resources by some of the alien counterparts, or the general reduction
of resources (habitat and food) owing to the collapse of the native
macrophyte meadows –or the combination of some or all stressors.
4.4 |Regional and global references, ecological
considerations, and implications for management and
conservation
The degraded reef state reported here characterizes all other locations
visited along the Israeli coast in the past 8 years (personal observations
and unpublished data from multiple surveys). As the rest of the south‐
eastern Levant is presumed to be exposed to similar pressures of
overfishing, bioinvasions and climate change, we suggest that these
findings are representative of a much larger area. One semi‐quantita-
tive study of the reefs around Kos (at the south‐eastern end of the
Aegean Sea, just outside of the area defined as the Levantine basin)
used identical survey methodologies in 1981 and 2013 to show that
in the past, the Kos reefs were dominated by flourishing shallow brown
algal (Cystoseira) forests but these reefs are mostly barrens now
(Bianchi, Corsini‐Foka, Morri, & Zenetos, 2014). Another study from
the Aegean, compared areas where rabbitfish are present (south, near
the Levant) and absent (north) (Vergés, Tomas, et al., 2014). It showed
that where rabbitfish are absent, the native herbivorous browser,
Sarpa salpa was abundant and algae forests flourish, and where
rabbitfish are present, S. salpa was scarce, canopy algae were 65% less
abundant, and there was a 60% lower overall benthic biomass and 40%
lower species richness. The few community surveys and several exper-
iments in southern Turkey (Sala et al., 2011, 2012) suggest that the
situation is very similar there as well, suggesting that the entire Levant
has gone through a profound ecosystem phase shift. This community
shift to barrens is part of a global phenomenon of tropicalization of
temperate reefs, where tropical herbivorous fish move into temperate
regions that are warming owing to climate change, causing regime
shifts of benthic communities, from algal forests to barrens, with
perhaps the most extreme case being the 'isoyake' phenomenon in
southern Japan, where kelp forests were replaced by deforested
barrens and then by coral reefs within two decades (Vergés, Steinberg,
et al., 2014; Vergés et al., 2016).
From an ecosystem functions and food web perspective, and
assuming that indeed our findings represent most reef areas of the
south‐eastern Levant, we suggest that the reef ecosystem in the
region has ecologically lost its natural macro‐herbivores −the fish,
Sarpa salpa and Sparisoma cretense, as well as sea urchins (although
all can still be found but in tiny numbers in a few locations along the
coast). It, however, ‘gained’highly effective invasive herbivores, the
two Indo‐Pacific rabbitfish, that appear to graze the native macroalgal
communities down to turf, as has been shown experimentally on the
Turkish Mediterranean coast (Sala et al., 2011; Vergés, Tomas, et al.,
2014) and also in Israel (Yeruham, Abelson, Spiegel, Rilov, unpublished
data). As suggested by Sala et al. (2011), the decimation of erect
algal beds, which are important settlement habitat for numerous
rocky‐bottom organisms (Cheminée et al., 2013; García‐Rubies &
Zabala, 1990), and represent permanent habitat (and source of food)
for numerous epibiotic algae and invertebrates (Ballesteros, 1990;
Panayotidis, Orfanidis, & Tsiamis, 2007; Sales & Ballesteros, 2011),
may have considerable cascading effects on the whole reef community
including the loss of overall primary productivity on the reefs.
Whether some of the well‐established aliens possess traits similar
to the ones lost with the loss of the natives, and thus can ecologically
replace them, is another very important question in the context of
ecosystem functions. We do not know, for example, if in the past the
native spiny oyster Spondylus gaederopus (that is now completely
absent, Shabtay et al., 2014), was more, less or as abundant as the
alien S. spinosus (which is very similar in size and shape, and presum-
ably function) is today on the south‐eastern Mediterranean reefs.
Because these are large filter‐feeders, their abundance can affect the
diversity and function of both the plankton (by feeding on it) and
benthos (as ecosystem engineers and as contributors to the detrital
pool, from their faeces or pseudo‐faeces). Similarly, it is unknown if
the newly established alien macroalgae compensate for some of the
functions lost by siganid over‐grazing of brown algae. Possibly, these
alien tropical macroalgae can maintain the functions provided by
macroalgae on the reefs (i.e. as food and habitat) during the summer
because they can persist during this increasingly warm season, when
the brown algae stands crash even at their ‘refugia’non‐grazed sites.
These kinds of questions require detailed (multi‐stressor) inves-
tigations on the ecosystem‐level impacts –i.e. biodiversity, traits and
functions –caused by the transformation of the reef seascape from
native meadows to turf, and from turf to habitat‐forming aliens, within
the context of climate change. If indeed some invaders compensate
for ecosystem functions (structural or biogeochemical) lost with the
collapse of native species (Rilov, 2016), then those alien species should
perhaps be ‘embraced’by managers.
The fact that the proportion of alien fish was not lower inside the
reserve is perhaps disappointing from a conservation perspective, but,
as mentioned above, this was a prevalent pattern in other MPAs in the
Mediterranean (Guidetti et al., 2014), and thus a different result in the
hottest hotspot of invasions in the basin –the south‐eastern Levant –
was unlikely. From a biogeographic perspective, with the further
expansion of the Suez Canal (Galil et al., 2015) and continued climate
change, one can only expect further transformation of Levant reefs
as well as a ‘spread’of this ecosystem phase‐shift further west in the
Mediterranean. Complementary to the Sala et al. (2012) survey, this
study can serve as a very ‘shifted baseline’(sensu Jackson et al.,
2001) for monitoring of the current state of Levant reefs and be used
for comparisons with other Mediterranean regions.
Under such ecologically‐degraded conditions what can conserva-
tion management do? Clearly, better regulation of local stressors,
RILOV ET AL.15
mainly overfishing (using reserves), may aid in mitigating at least some
of the impacts. For example, protection of predators supplemented
by encouragement of fishers to actively remove, and the public to
consume, rabbitfish over indigenous predators, may help control
siganid populations, although reserves alone will not reduce the
invaders’abundance but rather increase it –as seen in this study.
Controlling siganid populations in turn could alleviate the pressure on
the native macrophytes, thus regaining some of the lost biomass and
habitat. Active removal (including for consumption) by professionals
and citizens is currently attempted in the case of the lionfish invasion
in the Caribbean Sea (Frazer, Jacoby, Edwards, Barry, & Manfrino,
2012) where it had profound ecological impacts on the local reef
communities (Albins, 2015), had, however, mixed results (Scyphers
et al., 2015). There were also anecdotal records of Red Sea lionfish in
the past in Israel (Golani & Sonin, 1992), but a real lionfish invasion
may be presently unfolding in the Levant with increasing sightings in
Cyprus (Kletou, Hall‐Spencer, & Kleitou, 2016), Lebanon (Azzurro &
Bariche, 2017) and also recently in Israel (N. Stern, personal communi-
cations, 20 August 2016). This represents a further ecological threat
that would add more pressure on those highly modified reefs. It is pos-
sible, however, that predation by lionfish could help control the
rabbitfish populations if they indeed feed on them. Which invasion,
the veteran herbivorous rabbitfish or the newcomer predatory lionfish,
is more detrimental for the functioning of the Mediterranean reef
ecosystems, and how would they interact, is yet to be seen. Obviously,
in this very complex ecological scenario; making sensible conservation
decisions is a similarly complex and challenging task.
Recent years have shown that Israel is making a genuine effort to
improve marine conservation in its Mediterranean territorial waters,
and comply, as best as possible, with European directives, mostly so,
with the Marine Strategy Framework Directive (MSFD), aimed to
advance Good Environmental Status (GES). A network of large and
small reserves, covering up to 20% of Israel's territorial waters has
been proposed by INPA (Yahel, 2010), much of it including subtidal
reefs. As part of this effort, the Achziv reserve has recently been signif-
icantly enlarged, while an INPA ranger has been appointed to protect
the small Shikmona reserve in Haifa (which was mostly a ‘paper park’
since its declaration in 2008). In addition, newly‐updated fishing
regulations were published in 2016 by the Fishery Department, includ-
ing a total ban of fishing by all sectors during the reproduction season
(with a promise of new funding for enforcement). These efforts will
hopefully reduce fishing pressure, increase the populations of reef
predators, which may in‐turn alleviate some of the grazing pressure
on canopy‐forming macroalgae by the siganids. There is also a govern-
ment move to expand the environmental monitoring program, which
should improve tracking of ecological changes and aid in assessing the
success of new conservation measures. Nevertheless, policy makers
and conservation managers must accept the fact that the Levant
ecosystems are at such a highly‐altered ecological state –even a
regime‐shift –that some of the MSFD targets are simply not realistic.
Biological invasions are here to stay; thus, GES descriptor 2 in the direc-
tive: ‘Non‐indigenous species do not adversely alter the ecosystem’is a
goal unfortunately not achievable. In addition, the rapid ocean warming
(Rilov, 2016) enhances the tropicalization of the region, and thus,
conservation goals and strategies must be more adaptive. For example,
a more realistic conservation target might be the preservation or restora-
tion of ecosystem functions rather than the original native biodiversity.
These aspects have not been addressed in a recent list of recommenda-
tions for management of non‐indigenous species in marine ecosystems
(Ojaveer et al., 2014); they should, however, be recognized in future
conservation planning, so that more realistic goals can be defined.
ACKNOWLEDGEMENTS
We thank IOLR's Marine Community Ecology Lab past research assis-
tants (A. Konstantinovsky, N. Bartov, E. Israeli, Y. Hyams, U. Arkin, A.
Barash, M. Grossowicz, U. Zacharia, O. Almog, O. Frid, O. Abramson)
for their extensive efforts in both field (40–50 dives per season) and
lab work. We thank the marine crew of IOLR for their help in the field
(especially E. Hagai). Our study in the marine reserve was greatly
supported by EcoOcean that provided their research boat under
reduced cost, as well as the Israel Parks and Nature Authority that
helped with the logistics and with post‐survey site mapping. Several
people helped with taxonomic identification including A. Israel with
macroalgae, H. Mienis with mollusca, S. Shefer with sponges and E.
Mizrahi with worms. The Rinkevich lab at IOLR did the DNA barcoding
and identified several of the live rock specimens for the Haifa
surveys. This work was supported mainly by the Goldman Foundation,
the Israel Port Authority, and also partly by grants from the Marie
Curie Reintegration Programme under the European Union's Seventh
Framework, grant number 249147 (to G.R.), and the Israel Science
Foundation, grant number 1217/10 (to G.R.).
CONFLICT OF INTEREST
There are no conflicts of interest
ORCID
Gil Rilov http://orcid.org/0000-0002-1334-4887
REFERENCES
Agnetta, D., Badalamenti, F., Ceccherelli, G., Di Trapani, F., Bonaviri, C., &
Gianguzza, P. (2015). Role of two co‐occurring Mediterranean sea
urchins in the formation of barren from Cystoseira canopy. Estuarine,
Coastal and Shelf Science,152,73–77.
Albins, M. A. (2015). Invasive Pacific lionfish Pterois volitans reduce abun-
dance and species richness of native Bahamian coral‐reef fishes.
Marine Ecology Progress Series,522, 231–243.
Anderson, M. J., Gorley, R. N., & Clarke, K. R. (2008). PERMANOVA+ for
PRIMER: Guide to software and statistical methods. Plymouth, UK:
PRIMER‐E, Ltd.
Arenas, F., Sánchez, I., Hawkins, S. J., & Jenkins, S. R. (2006). The invasibility
of marine algal assemblages: Role of functional diversity and identity.
Ecology,87, 2851–2861.
Aronov, A., & Goren, M. (2008). Ecology of the mottled grouper
(Mycteroperca rubra) in the eastern Mediterranean. Electronic Journal
of Ichthyology,2,43–55.
Aronson, R. B., Bruno, J. F., Precht, W. F., Glynn, P. W., Harvell, C. D.,
Kaufman, L., …Valentine, J. F. (2003). Causes of coral reef degradation.
Science,302, 1502–1502.
Azzurro, E., & Bariche, M. (2017). Local knowledge and awareness on the
incipient lionfish invasion in the eastern Mediterranean Sea. Marine
and Freshwater Research,68(10), 1950–1954.
16 RILOV ET AL.
Ballesteros, E. (1990). Structure and dynamics of the community of
Cystoseira zosteroides (turner) C. Agardh (Fucales, Phaeophyceae) in
the northwestern Mediterranean. Scientia Marina,54, 217–299.
Barash, A., & Danin, Z. (1982). Mediterranean mollusca of Israel and Sinai:
Composition and distribution. Israel Journal of Zoology,31,86–118.
Ben‐Tuvia, A. (1964). Two siganid fishes of Red Sea origin in the eastern
Mediterranean. In Bulletin. Sea Fisheries Research Station (Haifa)
(pp. 1–9). . Haifa: Ministry of Agriculture. Department of Fisheries.
Sea Fisheries Research Station.
Bianchi, C., Corsini‐Foka, M., Morri, C., & Zenetos, A. (2014). Thirty years
after‐dramatic change in the coastal marine habitats of Kos Island
(Greece), 1981‐2013. Mediterranean Marine Science,15, 482–497.
Bitar, G., Ramos‐Esplá, A., Ocaña, O., Sghaier, Y., Forcada, A., Valle, C., …
Verlaque, M. (2017). Introduced marine macroflora of Lebanon and its
distribution on the Levantine coast. Mediterranean Marine Science,18,
138–155.
Canning‐Clode, J., Maloney, K. O., McMahon, S. M., & Wahl, M. (2010).
Expanded view of the local‐regional richness relationship by incorporat-
ing functional richness and time: A large‐scale perspective. Global
Ecology and Biogeography,19, 875–885.
Cardona, L., Sales, M., & Lopez, D. (2007). Changes in fish abundance do
not cascade to sea urchins and erect algae in one of the most oligotro-
phic parts of the Mediterranean. Estuarine, Coastal and Shelf Science,72,
273–282.
Cheminée, A., Sala, E., Pastor, J., Bodilis, P., Thiriet, P., Mangialajo, L., …
Francour, P. (2013). Nursery value of Cystoseira forests for Mediterra-
nean rocky reef fishes. Journal of Experimental Marine Biology and
Ecology,442,70–79.
Coll, M., Piroddi, C., Albouy, C., Lasram, F. B., Cheung, W. W. L.,
Christensen, V., …Pauly, D. (2012). The Mediterranean Sea under siege:
Spatial overlap between marine biodiversity, cumulative threats and
marine reserves. Global Ecology and Biogeography,21, 465–480.
Coll, M., Piroddi, C., Steenbeek, J., Kaschner, K., Lasram, F. B., Aguzzi, J., …
Voultsiadou, E. (2010). The biodiversity of the Mediterranean Sea:
Estimates, patterns, and threats. PLoS ONE,5.
Connell, S., Foster, M., & Airoldi, L. (2014). What are algal turfs? Towards a
better description of turfs. Marine Ecology Progress Series,495,299–307.
Crocetta, F., & Russo, P. (2013). The alien spreading of Chama pacifica
Broderip, 1835 (Mollusca: Bivalvia: Chamidae) in the Mediterranean
Sea. Turkish Journal of Zoology,37,92–96.
Diamant, A., Bentuvia, A., Baranes, A., & Golani, D. (1986). An analysis of
rocky coastal eastern Mediterranean fish assemblages and a compari-
son with an adjacent small artificial reef. Journal of Experimental
Marine Biology and Ecology,97, 269–285.
Downie, R. A., Babcock, R. C., Thomson, D. P., & Vanderklift, M. A. (2013).
Density of herbivorous fish and intensity of herbivory are influenced by
proximity to coral reefs. Marine Ecology Progress Series,482, 217–225.
Edelist, D., Rilov, G., Golani, D., Carleton, J., & Spanier, E. (2013).
Restructuring the sea: Profound shifts in the world's most invaded
marine ecosystem. Diversity and Distributions,19,69–77.
Edelist, D., Scheinin, A., Sonin, O., Shapiro, J., Salameh, P., Rilov, G., …Zeller,
D. (2013). Israel: Reconstructed estimates of total fisheries removals in
the Mediterranean, 1950‐2010. Acta Adriatica,54, 253–263.
Edelstein, T. (1960). The biology and ecology of deep sea algae of the Haifa
Bay area. PhD thesis, Hebrew University, Jerusalem.
Falkenberg, L. J., Connell, S. D., Coffee, O. I., Ghedini, G., & Russell, B. D.
(2015). Species interactions can maintain resistance of subtidal algal
habitats to an increasingly modified world. Global Ecology and Conserva-
tion,4, 549–558.
Fanelli,E.,Azzurro,E.,Bariche,M.,Cartes,J.E.,&Maynou,F.(2015).Depicting
the novel eastern Mediterranean food web: A stable isotopes study
following Lessepsian fish invasion. Biological Invasions,17, 2163–2178.
Ferrer, E., Garreta, G., & Ribera, M. A. (1997). Effect of Caulerpa taxifolia on
the productivity of two Mediterranean macrophytes. Marine Ecology
Progress Series,149, 279–287.
Frazer, T. K., Jacoby, C. A., Edwards, M. A., Barry, S. C., & Manfrino, C. M.
(2012). Coping with the lionfish invasion: Can targeted removals yield
beneficial effects? Reviews in Fisheries Science,20, 185–191.
Galil, B. S., Boero, F., Campbell, M. L., Carlton, J. T., Cook, E., Fraschetti, S.,
…Macpherson, E. (2015). Double trouble’: The expansion of the Suez
Canal and marine bioinvasions in the Mediterranean Sea. Biological
Invasions,17, 973–976.
García‐Rubies, A., & Zabala, M. (1990). Effects of total fishing prohibition
on the rocky fish assemblages of Medes Islands marine reserve (NW
Mediterranean). Scientia Marina,54, 317–328.
Garrabou, J., Coma, R., Bensoussan, N., Bally, M., Chevaldonne, P., Cigliano,
M., …Cerrano, C. (2009). Mass mortality in Northwestern Mediterra-
nean rocky benthic communities: Effects of the 2003 heat wave.
Global Change Biology,15, 1090–1103.
Gattuso, J.‐P., Magnan, A., Billé, R., Cheung, W., Howes, E., Joos, F., …Eakin,
C. (2015). Contrasting futures for ocean and society from different
anthropogenic CO
2
emissions scenarios. Science,349,4722‐1–4722‐10.
Giakoumi, S., Cebrian, E., Kokkoris, G. D., Ballesteros, E., & Sala, E. (2012).
Relationships between fish, sea urchins and macroalgae: The structure
of shallow rocky sublittoral communities in the Cyclades, Eastern Med-
iterranean. Estuarine, Coastal and Shelf Science,109,1–10.
Gofas, S., & Zenetos, A. (2003). Exotic molluscs in the Mediterranean basin:
Current status and perspectives. Oceanography and Marine Biology,41,
237–277.
Golani, D., & Sonin, O. (1992). New records of the Red Sea fishes, Pterois
miles (Scorpaenidae) and Pteragogus pelycus (Labridae) from the eastern
Mediterranean Sea. Japanese Journal of Ichthyology,39, 167–169.
Goren, M., & Galil, B. S. (2001). Fish biodiversity in the vermetid reef of
Shiqmona (Israel). Marine Ecology‐Pubblicazioni Della Stazione Zoologica
Di Napoli I,22, 369–378.
Goren, M., Lipsky, G., Brokovich, E., & Abelson, A. (2010). A ‘flood’of
alien cardinal fishes in the eastern Mediterranean‐first record of the
indo‐Pacific Cheilodipterus novemstriatus (Rüppell, 1838) in the
Mediterranean Sea. Aquatic Invasions,5, S49–S51.
Guidetti, P. (2005). Reserve effect and trophic cascades in Mediterranean
sublittoral rocky habitats: A case study at the marine protected area
of Torre Guaceto (southern Adriatic Sea). Biologia Marina Mediterranea,
12,99–105.
Guidetti, P. (2006). Marine reserves reestablish lost predatory interactions
and cause community changes in rocky reefs. Ecological Applications,
16, 963–976.
Guidetti, P., Baiata, P., Ballesteros, E., Di Franco, A., Hereu, B., Macpherson,
E., …Sala, E. (2014). Large‐scale assessment of Mediterranean Marine
Protected Areas effects on fish assemblages. PLoS ONE,9(4), 1–14.
e91841.
Guidetti, P., & Dulcic, J. (2007). Relationships among predatory fish, sea
urchins and barrens in Mediterranean rocky reefs across a latitudinal
gradient. Marine Environmental Research,63, 168–184.
Guidetti, P., & Sala, E. (2007). Community‐wide effects of marine reserves
in the Mediterranean Sea. Marine Ecology‐Progress Series,335,43–56.
Guy Haim, T. (2017). The impact of ocean warming and acidification on
coastal benthic species and communities. PhD thesis, University of
Haifa, Haifa.
Halpern, B. S., Frazier, M., Potapenko, J., Casey, K. S., Koenig, K., Longo, C.,
…Selkoe, K. A. (2015). Spatial and temporal changes in cumulative
human impacts on the world's ocean. Nature Communications,6.
https://doi.org/10.1038/ncomms8615
Harmelin‐Vivien, M. L., Bitar, G., Harmelin, J. G., & Monestiez, P. (2005).
The littoral fish community of the Lebanese rocky coast (eastern
Mediterranean Sea) with emphasis on Red Sea immigrants. Biological
Invasions,7, 625–637.
Herut, B. (2016). The National Monitoring Program of Israel's Mediterra-
nean waters ‐Scientific Report for 2015, IOLR, Report H42/2016.
National Institute of Oceanography, Israel Oceanographic and Limno-
logical Research.
RILOV ET AL.17
Hoffman, R., Israel, A., Lipkin, Y., Dubinsky, Z., & Iluz, D. (2008). First record
of two seaweeds from the Israeli Mediterranean: Galaxaura rugosa (J.
Ellis and Solander) JV Lamouroux (Rhodophyta) and Codium adhaerens
C. Agardh (Chlorophyta). Israel Journal of Plant Sciences,56, 123–126.
Israel, A., Einav, R., Silva, P. C., Paz, G., Chacana, M. E., & Douek, J. (2010).
First record of the seaweed Codium parvulum (Chlorophyta) in the Med-
iterranean waters: Recent blooms on the northern shores of Israel.
Phycologia,49(2), 107–112.
Jackson, J. B. C. (2008). Ecological extinction and evolution in the brave
new ocean. Proceedings of the National Academy of Sciences of the
United States of America,105, 11458–11465.
Jackson, J. B. C., Kirby, M. X., Berger, W. H., Bjorndal, K. A., Botsford, L. W.,
Bourque, B. J., …Warner, R. R. (2001). Historical overfishing and the
recent collapse of coastal ecosystems. Science,293, 629–638.
Jouffray, J.‐B., Nyström, M., Norström, A. V., Williams, I. D., Wedding, L. M.,
Kittinger, J. N., & Williams, G. J. (2015). Identifying multiple coral reef
regimes and their drivers across the Hawaiian archipelago. Philosophical
Transactions of the Royal Society, B: Biological Sciences,370, 20130268.
Katsanevakis, S., Coll, M., Piroddi, C., Steenbeek, J., Ben Rais Lasram, F.,
Zenetos, A., & Cardoso, A. C. (2014). Invading the Mediterranean Sea:
Biodiversity patterns shaped by human activities. Frontiers in Marine
Science, 1. https://doi.org/10.3389/fmars.2014.00032
Klein, J., & Verlaque, M. (2008). The Caulerpa racemosa invasion: A critical
review. Marine Pollution Bulletin,56, 205–225.
Kletou, D., Hall‐Spencer, J. M., & Kleitou, P. (2016). A lionfish (Pterois miles)
invasion has begun in the Mediterranean Sea. Marine Biodiversity
Records,9,1.
Kohler, K. E., & Gill, S. M. (2006). Coral point count with excel extensions
(CPCe): A visual basic program for the determination of coral and sub-
strate coverage using random point count methodology. Computers &
Geosciences,32, 1259–1269.
Lejeusne, C., Chevaldonné, P., Pergent‐Martini, C., Boudouresque, C. F., &
Pérez, T. (2010). Climate change effects on a miniature ocean: The
highly diverse, highly impacted Mediterranean Sea. Trends in Ecology
& Evolution,25, 250–260.
Lobel, P. S. (2005). Scuba Bubble noise and fish behavior: A rationale for
silent diving technology.
Lundberg, B. (1996) Composition of the seaweed vegetation along the
Mediterranean coast of Israel. Jerusalem, Nature Reserves Authority.
Madin, E. M., Madin, J. S., & Booth, D. J. (2011). Landscape of fear visible
from space. Scientific Reports,1. https://doi.org/10.1038/srep00014
Marbà, N., Jorda, G., Agusti, S., Girard, C., & Duarte, C. M. (2015). Foot-
prints of climate change on Mediterranean Sea biota. Frontiers in
Marine Science,2. https://doi.org/10.3389/fmars.2015.00056
Mienis, H., Zaslow, R., & Rittner, O. (2012). First records of Alectryonella
plicatula from the Mediterranean coast of Israel (Mollusca, Bivalvia,
Ostreidae). Triton,26,4–5.
Mienis, H. K., Galili, E., & Rapoport, J. (1993). The spiny oyster, Spondylus
spinosus, a well‐established indo‐Pacific bivalve in the eastern Mediter-
ranean off Israel (Mollusca, Bivalvia, Spondylidae). Zoology in the Middle
East,9,83–92.
Nawrot, R., Chattopadhyay, D., & Zuschin, M. (2015). What guides invasion
success? Ecological correlates of arrival, establishment and spread of
Red Sea bivalves in the Mediterranean Sea. Diversity and Distributions,
21, 1075–1086.
Nicolay, K., & Manoja, E. R. (1983). Strombus (Conomurex) decorus raybaudii
n. spp. La Conchiglia,14,17–18.
Norton, T. (1992). Dispersal by macroalgae. British Phycological Journal,27,
293–301.
Nunez‐Lara, E., Arias‐Gonzalez, J. E., & Legendre, P. (2005). Spatial patterns
of Yucatan reef fish communities: Testing models using a multi‐scale
survey design. Journal of Experimental Marine Biology and Ecology,324,
157–169.
Ojaveer, H., Galil, B. S., Minchin, D., Olenin, S., Amorim, A., Canning‐Clode,
J., …Zenetos, A. (2014). Ten recommendations for advancing the
assessment and management of non‐indigenous species in marine
ecosystems. Marine Policy,44, 160–165.
Panayotidis, P., Orfanidis, S., & Tsiamis, K. (2007). Cystoseira crinita commu-
nity in the Aegean Sea. Rapport Commission International pour
l'exploration scientifique de la Mer Mediterranee,38, 570.
Peleg, O. (2016). Bioinvasions drive major shifts in Levant reef community
structure and ecosystem functioning. MSc thesis, University of Haifa,
Haifa.
Piazzi, L., Ceccherelli, G., & Cinelli, F. (2001). Threat to macroalgal diversity:
Effects of the introduced green alga Caulerpa racemosa in the Mediter-
ranean. Marine Ecology Progress Series,210, 149–159.
Pinnegar, J. K., Polunin, N. V. C., Francour, P., Badalamenti, F., Chemello, R.,
Harmelin‐Vivien, M. L., …Pipitone, C. (2000). Trophic cascades in
benthic marine ecosystems: Lessons for fisheries and protected‐area
management. Environmental Conservation,27, 179–200.
Poloczanska, E. S., Brown, C. J., Sydeman, W. J., Kiessling, W., Schoeman,
D. S., Moore, P. J., …Burrows, M. T. (2013). Global imprint of climate
change on marine life. Nature Climate Change,3, 919–925.
Rilov, G. (2016). Multi‐species collapses at the warm edge of a warming sea.
Scientific Reports,6. https://doi.org/10.1038/srep36897
Rilov, G., & Galil, B. (2009). Marine bioinvasions in the Mediterranean
Sea ‐history, distribution and ecology. In G. Rilov, & J. A. Crooks
(Eds.), Biological invasions in marine ecosystems: Ecological, management,
and geographic perspectives (pp. 549–575). Heidelberg, Germany:
Springer‐Verlag.
Safriel, U. N. (1975). The role of vermetid gastropods in the formation of
Mediterranean and Atlantic reefs. Oecologia (Berlin),20,85–101.
Sala, E., Ballesteros, E., Dendrinos, P., Di Franco, A., Ferretti, F., Foley, D., …
Zabala, M. (2012). The structure of Mediterranean rocky reef ecosys-
tems across environmental and human gradients, and conservation
implications. PLoS ONE,7. https://doi.org/10.1371/journal.pone.0032742
Sala, E., Boudouresque, C. F., & Harmelin‐Vivien, M. (1998). Fishing, trophic
cascades, and the structure of algal assemblages: Evaluation of an old
but untested paradigm. Oikos,82, 425–439.
Sala, E., Kizilkaya, Z., Yildirim, D., & Ballesteros, E. (2011). Alien marine
fishes deplete algal biomass in the eastern Mediterranean. PLoS ONE,
6. https://doi.org/10.1371/journal.pone.0017356
Sala, E., & Zabala, M. (1996). Fish predation and the structure of the sea
urchin Paracentrotus lividus populations in the NW Mediterranean.
Marine Ecology‐Progress Series,140,71–81.
Sales, M., & Ballesteros, E. (2011). Seasonal dynamics and annual production
of Cystoseira crinita (Fucales: Ochrophyta)‐dominated assemblages
from the northwestern Mediterranean. Scientia Marina,76, 391–401.
Scyphers, S. B., Powers, S. P., Akins, J. L., Drymon, J. M., Martin, C. W.,
Schobernd, Z. H., …Switzer, T. S. (2015). The role of citizens in de-
tecting and responding to a rapid marine invasion. Conservation
Letters,8, 242–250.
Seber, G. A. F. (1982). The estimation of animal abundance. London, UK:
Griffin.
Shabtay, A., Tikochinski, Y., Benayahu, Y., & Rilov, G. (2014). Preliminary
data on the genetic structure of a highly successful invading population
of oyster suggesting its establishment dynamics in the Levant. Marine
Biology Research,10, 407–415.
Shaltout, M., & Omstedt, A. (2014). Recent sea surface temperature
trends and future scenarios for the Mediterranean Sea. Oceanologia,
56, 411–443.
Sharon, Y., Benayahu, Y., & Mienis, H. (2005). First record of an exotic
oyster: Alectryonella crenulifera, from the Mediterranean coast of Israel.
Triton,12,5–6.
Sisma‐Ventura, G., Yam, R., & Shemesh, A. (2014). Recent unprecedented
warming and oligotrophy of the eastern Mediterranean Sea within the
last millennium. Geophysical Research Letters,41, 5158–5166.
Steffen, W., Crutzen, P. J., & McNeill, J. R. (2007). The Anthropocene: Are
humans now overwhelming the great forces of nature. Ambio: A Journal
of the Human Environment,36, 614–621.
18 RILOV ET AL.
Steinitz, W. (1927). Beiträge zur Kenntnis der Küstenfauna Palästinas. Part
1 of Publicatione dell Stazione Zoologica di Napoli, 8, 331‐353.
Steneck, R. S., Graham, M. H., Bourque, B. J., Corbett, D., Erlandson,
J. M., Estes, J. A., & Tegner, M. J. (2002). Kelp forest ecosystems:
Biodiversity, stability, resilience and future. Environmental Conserva-
tion,29, 436–459.
Vergés, A., Doropoulos, C., Malcolm, H. A., Skye, M., Garcia‐Pizá, M.,
Marzinelli, E. M., …Vila‐Concejo, A. (2016). Long‐term empirical
evidence of ocean warming leading to tropicalization of fish
communities, increased herbivory, and loss of kelp. Proceedings of the
National Academy of Sciences of the United States of America,113(48),
13791–13796, 201610725.
Vergés, A., Steinberg, P. D., Hay, M. E., Poore, A. G., Campbell, A. H.,
Ballesteros, E., …Feary, D. A. (2014). The tropicalization of temperate
marine ecosystems: Climate‐mediated changes in herbivory and com-
munity phase shifts. Proceedings of the Royal Society B: Biological
Sciences,281, 20140846.
Vergés, A., Tomas, F., Cebrian, E., Ballesteros, E., Kizilkaya, Z., Dendrinos, P.,
…Sala, E. (2014). Tropical rabbitfish and the deforestation of a warming
temperate sea. Journal of Ecology,102, 1518–1527.
Wahl, M., Buchholz, B., Winde, V., Golomb, D., Guy‐Haims,T., Müller, J. D., …
Böttcher, M. E. (2015). A mesocosm concept for the simulation of near‐
natural shallow underwater climates: The Kiel outdoor Benthocosms
(KOB). Limnology and Oceanography: Methods,13,651–663.
Watson, D. L., & Harvey, E. S. (2007). Behaviour of temperate and sub‐
tropical reef fishes towards a stationary SCUBA diver. Marine and
Freshwater Behaviour and Physiology,40,85–103.
Wernberg, T., Smale, D. A., Tuya, F., Thomsen, M. S., Langlois, T. J., de
Bettignies, T., …Rousseaux, C. S. (2013). An extreme climatic event
alters marine ecosystem structure in a global biodiversity hotspot.
Nature Climate Change,3,78–82.
Wright, J. T., McKenzie, L. A., & Gribben, P. E. (2007). A decline in the abun-
dance and condition of a native bivalve associated with Caulerpa
taxifolia invasion. Marine and Freshwater Research,58, 263–272.
Yahel, R. (2010). Nature conservation policy in the Mediterranean Sea: Marine
nature reserves as a tool for environmental and biodiversity protection in
the Mediterranean Sea (in Hebrew). Israel: Israel Nature and Parks
Authority.
Yeruham, E. (2013). Possible explanations for Paracentrotus lividus
(European purple sea urchin) population collapse in South‐East
Mediterranean. MSc thesis, Tel Aviv University, Tel Aviv.
Yeruham, E., Rilov, G., Shpigel, M., & Abelson, A. (2015). Collapse of the
echinoid Paracentrotus lividus populations in the eastern Mediterranean
‐result of climate change? Scientific Reports,5, 13479.
Zenetos, A., Ovalis, P., & Kalogirou, S. (2009). Closing the gap: Cerithium
scabridum Philippi, 1848 found in the south Aegean (Greece,
Mediterranean Sea). Journal of Biological Research‐Thessaloniki,11,
107–110.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
supporting information tab for this article.
How to cite this article: Rilov G, Peleg O, Yeruham E, Garval
T, Vichik A, Raveh O. Alien turf: Overfishing, overgrazing and
invader domination in south‐eastern Levant reef ecosystems.
Aquatic Conserv: Mar Freshw Ecosyst. 2017;1–19. https://doi.
org/10.1002/aqc.2862
RILOV ET AL.19
