Journal of

Marine Science
and Engineering

Article
Comparative Study of Marine Cave Communities in a Protected
Area of the South-Eastern Aegean Sea, Greece
Markos Digenis 1,2, * , Christos Arvanitidis 2,3 , Thanos Dailianis 2 and Vasilis Gerovasileiou 1,2, *

1 Department of Environment, Faculty of Environment, Ionian University, 29100 Zakynthos, Greece

2 Hellenic Centre for Marine Research (HCMR), Institute of Marine Biology, Biotechnology and
Aquaculture (IMBBC), 71500 Heraklion, Greece; [email protected] (C.A.); [email protected] (T.D.)
3 LifeWatch ERIC, Sector II-III, Plaza de España, 41071 Seville, Spain
* Correspondence: [email protected] (M.D.); [email protected] (V.G.)

Abstract: Although more than 600 marine caves have been recorded so far along the Greek coasts
of the Aegean Sea (Eastern Mediterranean), only a few have been systematically studied for their
biodiversity. In this study, the benthic communities of six marine caves within a Protected Area of
South-Eastern Aegean were studied for the first time, both qualitatively and quantitatively. The
association of different geomorphological and topographical factors of the caves with the benthic
community structure was investigated. A total of 120 photographic quadrats covering the entrance
and semi-dark cave zones were analysed, with regard to coverage and taxon abundance, while
motile taxa were qualitatively recorded by visual census. The ecological quality status of the caves
was also assessed under an ecosystem-based approach. In total, 81 sessile and 45 motile taxa were
recorded, including 12 protected and 10 non-indigenous species. Multivariate community analysis
demonstrated that the geomorphological and topographical variables of the caves are significantly
associated with the observed biotic patterns. The ecological quality of the caves was assessed as poor
or moderate according to the CavEBQI index, highlighting the necessity for systematic monitoring.
Citation: Digenis, M.; Arvanitidis, C.; This study paves the way for similar studies in marine cave habitats aiming at the development of
Dailianis, T.; Gerovasileiou, V. management and conservation actions.
Comparative Study of Marine Cave
Communities in a Protected Area of Keywords: dark habitats; benthic communities; photoquadrats; visual census; scientific diving;
the South-Eastern Aegean Sea, Marine Protected Area; Eastern Mediterranean; non-indigenous species; habitat pressures; ecological
Greece. J. Mar. Sci. Eng. 2022, 10, 660. quality
https://doi.org/10.3390/jmse10050660

Academic Editor: Jean-Claude

Dauvin
1. Introduction
Received: 18 April 2022
Accepted: 10 May 2022
Mediterranean marine caves host rich biodiversity, including more than 2300 taxa, and
Published: 13 May 2022
provide shelter to several endemic, rare, and protected species [1,2]. They are characterised
by steep environmental gradients [3] which generate a characteristic zonation of benthic
Publisher’s Note: MDPI stays neutral
communities succeeding one another on a scale of few meters [2–4]. Such patterns have
with regard to jurisdictional claims in
been mainly studied in a qualitative manner, focusing specifically on sponges, which are the
published maps and institutional affil-
dominant colonisers on cave walls [5–7] while fewer studies involve quantitative analyses
iations.
of benthic community structure [4,8–10].
Marine caves are characterized by a rich geomorphological variety. According to
Riedl [3], six cave types exist in the Mediterranean Sea, depending on their level of submer-
Copyright: © 2022 by the authors.
sion (submerged or semi-submerged), shape and number of openings (e.g., blind-ended
Licensee MDPI, Basel, Switzerland. or tunnel-shaped). Geomorphological and topographic features such as the cave depth
This article is an open access article and orientation, entrance area and cave type, affect the level of light diffusion and water
distributed under the terms and renewal inside marine caves [3,7]. Such abiotic gradients also affect the structure of hard
conditions of the Creative Commons substrate benthic communities, resulting in biotic heterogeneity even among neighbouring
Attribution (CC BY) license (https:// marine caves, also known as “individuality” of each cave [2–4,9,11].
creativecommons.org/licenses/by/ Marine caves constitute a typical ecosystem of the Mediterranean Sea and are protected
4.0/). by the European Union’s Habitats Directive (92/43/EEC) and the Barcelona Convention

J. Mar. Sci. Eng. 2022, 10, 660. https://doi.org/10.3390/jmse10050660 https://www.mdpi.com/journal/jmse

J. Mar. Sci. Eng. 2022, 10, 660 2 of 16

under the Dark Habitats Action Plan [12]. So far, more than 3000 marine caves have been
recorded across the Mediterranean rocky coastline [2,13], 622 of which are located on the
Greek coasts of the Aegean Sea [14]. Although marine caves are known to be located
in 33 out of 62 Mediterranean Marine Protected Areas (MPAs) [15], the exact number of
marine caves in MPAs remains unknown. No specific management plans or regulations
are implemented for marine caves in most countries [1,2,16].
Over the last years, various impacts have been detected in marine caves, such as the
effects of climate change on their biota [17–21], coastal construction and marine pollu-
tion [13,22–24], as well as unregulated recreational activities [13,25,26]. In addition, more
than 60 non-indigenous species (NIS) and cryptogenic taxa (i.e., those that cannot be reliably
demonstrated as being either introduced or native) have been recorded in Mediterranean
marine caves [27–30]. Particularly in the Eastern Mediterranean, marine caves may function
as “stepping stones” for the expansion of sciaphilic NIS, such as invertebrates and fish of
Indo-Pacific origin which shelter in caves and coral reef cavities within their natural range
of distribution [28].
The current study aims to provide a quantitative description of hard-substrate benthic
communities and a catalogue of motile taxa (including NIS) recorded in six marine caves in
an MPA of the Eastern Mediterranean Sea using non-destructive methods. These data were
used to assess the ecological status of the surveyed marine caves through an ecosystem-
based index for the first time in the Eastern Mediterranean basin. In addition, it examines
the association of benthic community patterns with distinctive geomorphological and
topographic features of the studied caves. Through this study, three hypotheses were
tested: (i) all studied caves are significantly differentiated mainly due to cave-specific
features (individuality hypothesis); (ii) the examined geomorphological and topographic
factors (cave type, entrance depth, entrance area, and entrance orientation) significantly
affect the benthic community patterns, and (iii) equivalent ecological zones from different
caves show higher similarity levels.

2. Materials and Methods

2.1. Study Area
Six marine caves were surveyed in October 2019 in the Marine Protected Area of North
Karpathos and Saria Islands in the South-Eastern Aegean Sea (Greece), which is included in
the Greek NATURA 2000 Network (code GR4210003). This Eastern Mediterranean marine
area was studied for the first time regarding its marine cave biota. The six surveyed marine
caves were selected according to representativeness criteria [12] among 76 cave formations
recorded within the MPA during an extended exploration covering 45 km of coastline [31].
Out of the six caves examined, two were fully submerged, located at the north-eastern coast
of Karpathos Island, while four semi-submerged caves were located along the northeast to
western coastline of Saria Island (Figure 1, Table 1). The maximum seafloor depth of the
examined caves ranged from 4 to 11 m for fully submerged and 9–16 m for semi-submerged
caves, while the cave length ranged from 16 to 37 m. All examined caves were blind-ended
(i.e., had a single opening/entrance).

Table 1. Geomorphological and topographic characteristics of the six studied marine caves. Sub.,
fully submerged; Semi-sub., semi-submerged.

Latitude Longitude Entrance Entrance

Cave Name Type Depth (m) Length (m)
(DD) (DD) Width (m) Orientation
Alona 35.81969 27.23049 Sub. 1–4 24 11 E
Troulakas 35.81072 27.23314 Sub. 8–11 16 13 E
Giourious 35.86615 27.20269 Semi-sub. 0–14 17 10 W
Oxonisos 35.90087 27.21935 Semi-sub. 0–16 17 10 N
Palatia 35.88730 27.23276 Semi-sub. 0–9 17 4 S
Panteleimonas 35.85458 27.19273 Semi-sub. 0–15 37 30 W
 Mar. Sci. Eng. 2022, 10,
J. Mar. 9, x660
FOR PEER REVIEW 3 of 16

Figure 1. Location
Figure 1. Location of the studied
of the studied marine
marine caves
caves on the north
on the north coasts
coasts of
of Karpathos
Karpathos and
and Saria
Saria Islands
Islands
(A) and their location in Greece, Eastern Mediterranean (B).
(A) and their location in Greece, Eastern Mediterranean (B).

2.2. Sampling
Table 1. Geomorphological and topographic characteristics of the six studied marine caves. Sub.,
fully submerged; Semi-sub.,
SCUBA diving semi-submerged.
was used to approach the marine caves and perform the sampling.
For the study ofLatitude
hard substrate benthos sets of Depth
Longitude five random
Lengthstandard replicate
Entrance quadrats
Entrance
Cave
(25 Name
× 25 cm) were(DD)
photographed Type
(DD) on the opposite(m) walls of(m)
each cave
Widthzone(m)[10,32,33]. Pho-
Orientation
tographs
Alona were taken at a 4608
35.81969 27.23049by 3456 pixel
Sub. resolution
1–4 with a
24 digital camera
11 coupled
E with
two external underwater
Troulakas strobes. Cave
35.81072 27.23314 Sub.zones were
8–11 defined
16 according 13 to the bionomic
E
model developed
Giourious by Pérès 27.20269
35.86615 [34]: (i) entrance
Semi-sub.zone,0–14
which is usually
17 dominated
10 by sciaphilic
W
macroalgae
Oxonisos (e.g., rhodophytes),
35.90087 and Semi-sub.
27.21935 (ii) semi-dark zone, which
0–16 17 is dominated
10 by sciaphilic
N
animals, mostly 35.88730
Palatia sponges in27.23276
the Eastern Mediterranean
Semi-sub. 0–9 Sea [2,27].
17 None4of the studied S caves
had a distinct innermost
Panteleimonas 35.85458 completely
27.19273 dark zone, large
Semi-sub. 0–15 enough37 for proper30 sampling.W In total,
120 quadrats were collected from all caves. Additional close-up photos and qualitative
samples
2.2. were also collected for the identification of sessile taxa. Qualitative identification
Sampling
of motile
SCUBAspecies waswas
diving performed
used tovisually
approach withthe
a special
marinefocus onand
caves protected
performtaxa and
the NIS. Po-
sampling.
tential threats and pressures (e.g., necrosis of sessile taxa and litter) were
For the study of hard substrate benthos sets of five random standard replicate quadrats also documented
and×photographed,
(25 when present.on the opposite walls of each cave zone [10,32,33]. Pho-
25 cm) were photographed
tographs were taken at a 4608 by 3456 pixel resolution with a digital camera coupled with
2.3. Photoquadrats Processing
two external underwater strobes. Cave zones were defined according to the bionomic
model The percentage
developed of biotic
by Pérès [34]:cover for sessile
(i) entrance zone,benthos
which was calculated
is usually using by
dominated PhotoQuad,
sciaphilic
an advanced image processing software dedicated for marine biological applications
macroalgae (e.g., rhodophytes), and (ii) semi-dark zone, which is dominated by sciaphilic [35].
Each photoquadrat was analysed by overlaying 100 uniformly stratified points. Each point
animals, mostly sponges in the Eastern Mediterranean Sea [2,27]. None of the studied
was assigned to a sessile taxon or morpho-functional category (e.g., encrusting Rhodophyta,
caves had a distinct innermost completely dark zone, large enough for proper sampling.
turf-forming algae). Identification was held to the lowest feasible taxonomic level. Taxa
In total, 120 quadrats were collected from all caves. Additional close-up photos and qual-
that were present in the photoquadrats but did not fall below a random assignment point
itative samples were also collected for the identification of sessile taxa. Qualitative identi-
were given an arbitrary value of 0.5% cover [36–39]. Supplementary close-up photographic
fication of motile species was performed visually with a special focus on protected taxa
material was used to assist identification of sessile species. Percent coverage of the identified
and NIS. Potential threats and pressures (e.g., necrosis of sessile taxa and litter) were also
species for every quadrat was automatically calculated by PhotoQuad software [35].
documented and photographed, when present.
2.4. Ecological Quality Assessment
2.3. Photoquadrats Processing
The ecological quality index CavEBQI, as proposed by Rastorgueff et al. [22] for the
TheMediterranean,
Western percentage of biotic cover for
was applied sessilethe
to assess benthos was quality
ecological calculated
of allusing PhotoQuad,
the studied caves
an advanced image processing software dedicated for marine biological applications
in terms of ecosystem structure and functioning. The identified sessile taxa, whose coverage [35].
Each photoquadrat was analysed by overlaying 100 uniformly stratified
was calculated through photoquadrat analysis, were assigned to different components points. Each
point was assigned to a sessile taxon or morpho-functional category (e.g., encrusting
J. Mar. Sci. Eng. 2022, 10, 660 4 of 16

of the cave ecosystem as ‘Passive filter feeders’, ‘Large active filter feeders’, and ‘Small
active filter feeders’, while motile taxa were assigned as ‘Detritus feeders and omnivores’,
‘Characteristic carnivores’, and ‘Associate carnivores’ according to the model suggested
by Rastorgueff et al. [22]. Accidental cave visitors (e.g., herbivores) and NIS were not
considered in this assessment due to the lack of information regarding their habits in the
marine cave environment [27,28]. Sessile filter feeders’ stratification (with height of mm,
cm, or dm) and the presence of cave-dwelling mysids were also taken into account. The
Confidence Index of the Ecosystem-Based Quality Index (CIEBQI ), rescaled from 0 (worst
confidence in the value of the EBQI) to 10 (highest confidence in the value of the EBQI),
was also calculated for each studied cave [22].

2.5. Structure Assessment and Statistical Analysis

Multivariate, non-parametric resemblance analysis of the biotic coverage data was
performed through the software PRIMER-6 [40]. Coverage data were transformed under
the square root formula and a triangular similarity matrix was created based on the Bray-
Curtis similarity index [41]. Non-metric multidimensional scaling (nMDS) [42] was used
to investigate spatial patterns in the community structure. The association of geomorpho-
logical and topographic factor patterns with those of the benthic community was assessed
using one-way analysis of similarity (ANOSIM) [43] for six fixed factors: Cave (six levels,
one for each cave), Ecological Zone (two levels: Entrance, Semi-dark zone), Cave Type
(two levels, fully submerged and semi-submerged), Entrance Area (three levels: 15–40 m2 ,
110–230 m2 and 630 m2 ), Entrance Depth (max) (two levels: 0–10 m and 10–20 m) and
Entrance Orientation (four levels: west (W), east (E), north (N) and south (S)). The contribu-
tion of different taxa to Bray-Curtis dissimilarity among the resulting sample groups was
estimated through SIMPER (SIMilarity PERcentages) analysis [43].

3. Results
3.1. Sessile Community Structure
In total, 81 sessile taxa and morpho-functional categories were identified, classified
into Porifera (37), Bryozoa (13), Macroalgae (10), Cnidaria (6), Ascidiacea (6), Brachiopoda
(3), Mollusca (3), Foraminifera (1), Polychaeta (1), and Crustacea (1), listed in detail in Table
S1 in Supplementary Materials. Sixty-eight taxa were recorded from the cave entrance,
while 71 were recorded from the semi-dark zone. These taxa belonged to 47 species,
22 genera, 2 families, 2 orders, and 8 morpho-functional groups (Encrusting Rhodophyta,
Green filamentous algae, Turf-forming algae, Orange encrusting sponge, White sponge,
Yellow encrusting sponge, Bryozoan turf, Encrusting Bryozoa). The total number of taxa
varied among the different caves, from 43 in Alona and Palatia caves to 59 in Troulakas
cave (Oxonisos, Giourious, and Panteleimonas caves had 46, 48, and 50 taxa, respectively).
Eleven taxa were recorded only at the entrance zone, while 16 taxa were found exclusively
at the semi-dark zone of the studied caves (Table S1).
The percent coverage of sessile taxa was calculated separately for each ecological cave
zone (Tables 2 and S1, Figure 2). Macroalgae, dominated at the entrance zone (CE) of all
caves, reaching a maximum of 64% in the semi-submerged cave Palatia. Rhodophytes were
dominant at the cave entrance of all caves with lower percent coverage at the submerged
caves Troulakas and Alona (25% and 29%, respectively). Macroalgae covered up to 21% of
the semi-dark walls of Giourious cave.
Sponges dominated at the semi-dark zone of all caves (range of 16.9 to 59.9%), followed
by bryozoans, polychaetes and algae. For five out of the six examined caves, living sessile
communities did not occupy the full extent of the wall substrate at the semi-dark zone, as
manifested by the substantial percentages of non-living substrate (barren cave rock) in our
results (Figure 2). These percentages were high for Palatia and Alona (58.3 to 58.5%) and
lower for Oxonisos, Panteleimonas and Troulakas (14.7, 8.5, and 23.1%, respectively).
J. Mar. Sci. Eng. 2022, 10, 660 5 of 16

Table 2. Summary data with the average percentage of coverage (%) of all groups of sessile taxa and
non-living substrate for each ecological cave zone. ALO: Alona, GIOU: Giourious, OXO: Oxonisos,
PAL: Palatia, PANT: Panteleimonas, TROU: Troulakas, CE: entrance zone, SD: semi-dark zone, NLS:
Non-living substrate.

ALO GIOU OXO PAL PANT TROU

Taxa/Categories
CE SD CE SD CE SD CE SD CE SD CE SD
Macroalgae 28.5 - 56.2 20.9 47.6 8.6 64.2 1.1 64.1 10.4 24.7 0.7
Foraminifera 0.1 0.2 0.4 1.2 4.0 1.5 1.0 0.7 0.1 4.3 - 0.9
Porifera 27.2 16.9 11.7 52.9 30.8 53.2 4.0 26.9 6.9 59.9 16.4 53.2
J. Mar.Cnidaria
Sci. Eng. 2022, 9, x FOR
0.1 PEER REVIEW
- 0.5 1.0 - 1.0 - - 1.8 0.2 6.5 5 of 16
0.6
Polychaeta 5.6 10.6 0.2 0.2 3.0 3.0 1.4 11.1 0.3 0.4 15.7 8.3
Mollusca 0.3 1.0 - - - - - - - - - -
The percent coverage of sessile taxa was calculated separately for each ecological
Crustacea -
cave zone- (Tables-2 and S1,
- -
Figure-2). Macroalgae,
- -
mainly - 0.2
rhodophytes, -
dominated -
at the
Bryozoa 19.3 12.8
entrance29.9 22.7 of all16.9
zone (CE) 18.0
caves, reaching 10.2 1.9 of 64%
a maximum 25.9in the6.1 17.2
semi-submerged 13.2
cave
Ascidiacea 0.1 Palatia.
- Rhodophytes
0.1 - were - dominant
- at the
- cave entrance
- of all caves
0.2 - with lower
- percent
-
NLS 18.8 coverage
58.5 at
1.0 the submerged
1.1 4.0 caves
14.7Troulakas
20.1 and Alona
58.3 (25%
0.7 and
18.5 29%, respectively).
19.5 23.1
Macroalgae covered up to 21% of the semi-dark walls of Giourious cave.

Figure 2. Location of the studied caves and percent of coverage of identified sessile taxa and non-
Figure 2. Location of the studied caves and percent of coverage of identified sessile taxa and non-
living substrate is presented for each ecological cave zone (CE: cave entrance, SD: semi-dark zone).
living substrate is presented for each ecological cave zone (CE: cave entrance, SD: semi-dark zone).
Table 2. Summary data with the average percentage of coverage (%) of all groups of sessile taxa and
The total number of sessile taxa varied among each ecological zone, from a minimum
non-living substrate for each ecological cave zone. ALO: Alona, GIOU: Giourious, OXO: Oxonisos,
of 27 at the entrance zone of Oxonisos cave to a maximum of 40 at the entrance of Pan-
PAL: Palatia, PANT: Panteleimonas, TROU: Troulakas, CE: entrance zone, SD: semi-dark zone, NLS:
Non-living substrate.

ALO GIOU OXO PAL PANT TROU

Taxa/categories
CE SD CE SD CE SD CE SD CE SD CE SD
Macroalgae 28.5 - 56.2 20.9 47.6 8.6 64.2 1.1 64.1 10.4 24.7 0.7
 Sponges dominated at the semi-dark zone of all caves (range of 16.9 to 59.9%), fol-
lowed by bryozoans, polychaetes and algae. For five out of the six examined caves, living
sessile communities did not occupy the full extent of the wall substrate at the semi-dark
J. Mar. Sci. Eng. 2022, 10, 660 6 of 16
zone, as manifested by the substantial percentages of non-living substrate (barren cave
rock) in our results (Figure 2). These percentages were high for Palatia and Alona (58.3 to
58.5%) and lower for Oxonisos, Panteleimonas and Troulakas (14.7, 8.5, and 23.1%, respec-
teleimonas cave and from 17 to 44 at the semi-dark zone of Alona and Troulakas caves,
tively).
respectively
The total(Table
number3 and Tabletaxa
of sessile S1).varied
The total number
among of taxa increased
each ecological from
zone, from the cave
a minimum
of 27 at the entrance zone of Oxonisos cave to a maximum of 40 at the entrance ofcaves
entrance to the semi-dark zone for most caves except for Alona and Panteleimonas Pan-
(Table 3). The
teleimonas same
cave andpattern
from 17was
to observed
44 at the for Porifera,zone
semi-dark the group with
of Alona theTroulakas
and highest number
caves,
of taxa in total.
respectively (Tables 3 and S1). The total number of taxa increased from the cave entrance
to the 3.semi-dark
Table Summaryzone
datafor most
with the caves except for
total number Alona
of taxa by and Panteleimonas
sessile group for eachcaves (Tablecave
ecological 3).
The
zone.same
ALO:pattern was observed
Alona, GIOU: Giourious,forOXO:
Porifera, the group
Oxonisos, with the
PAL: Palatia, highest
PANT: number of taxa
Panteleimonas, TROU:in
total.
Troulakas, CE: entrance zone, SD: semi-dark zone.

ALO Table 3. Summary

GIOUdata with theOXOtotal number of taxa by sessile group
PAL for each ecological
PANT cave zone.
TROU
Taxa ALO: Alona, GIOU: Giourious, OXO: Oxonisos, PAL: Palatia, PANT: Panteleimonas, TROU: Trou-
CE SD CE: CE
lakas, entrance SD
zone, SD:CEsemi-darkSDzone. CE SD CE SD CE SD
Macroalgae 6 - 5 3 4 5 5 1 6 2 7 1
ALO GIOU OXO PAL PANT TROU
Foraminifera 1 1 Taxa1 1CE SD1 CE 1 SD 1
CE SD1 CE 1SD CE1 SD - CE 1
SD
Porifera 10 6 Macroalgae
15 166 -14 5 22 3 15 4 5 22 5 14 1 69 2 17 7 231
Cnidaria 2 1Foraminifera
1 41 11 1 4 1 -1 1 1 1 31 12 1 2 - 41
Porifera 10 6 15 16 14 22 15 22 14 9 17 23
Polychaeta 1 1 1 1 1 1 1 1 1 1 1 1
Cnidaria 2 1 1 4 1 4 - 1 3 2 2 4
Mollusca 2 1 Crustacea1 1- -- - - 1 -- - - - 1 - -- 1 - - --
Crustacea - - Polychaeta
- 11 1- 1 - 1 -1 1 - 1 - 1 11 1 - 1 -1
Bryozoa 7 3 Mollusca4 52 14 1 6 1 2- - 4 - 9 - 15 - 5 - 7-
Brachiopoda - 3 - 2 - - 2 2 1 - - 2
Brachiopoda - 3 - 2 - - 2 2 1 - - 2
Bryozoa 7 3 4 5 4 6 2 4 9 5 5 7
Ascidiacea 3 1 Ascidiacea
1 33 12 1 2 3 22 2 2 2 42 4- - 1 1 55
SUM 32 17 SUM 29 3732 1727 29 41 37 28
27 4134 28 4034 21
40 21 33 33 44
44

Among thetheidentified
identifiedsessile
sessiletaxa,
taxa,seven areare
seven protected
protectedaccording to the
according to Bern and Bar-
the Bern and
celona Conventions
Barcelona Conventionsandand
the the
Convention
Convention on on
International Trade
International in Endangered
Trade in Endangered Species of
Species
Wild Fauna
of Wild and
Fauna andFlora (CITES)
Flora (CITES)(Table
(TableS1).
S1).The
Thecircalittoral
circalittoralbivalve
bivalveNeopycnodonte
Neopycnodonte cochlear
was reported from the entrance and semi-dark zone of the Alona cave with 0.3% 0.3% and
and 1%
1%
coverage, respectively
respectively(Figure
(Figure3A,
3A,Table
TableS1). Bioconstructions
S1). Bioconstructions formed
formedby the
by serpulid pol-
the serpulid
ychaete Protula
polychaete tubularia
Protula (Figure
tubularia 3B) 3B)
(Figure as well as nodular
as well formations
as nodular of bryozoans
formations werewere
of bryozoans also
also recorded
recorded insemi-dark
in the the semi-dark
zonezone
of theof Panteleimonas
the Panteleimonasandand Oxonisos
Oxonisos caves,
caves, respectively.
respectively.

Figure
Figure 3.
3. The
Thebivalve
bivalveNeopycnodonte cochlear
Neopycnodonte (A),
cochlear serpulid
(A), bioconstructions
serpulid andand
bioconstructions a cardinal fish An-
a cardinal fish
thias anthias individual (B) photographed in the semi-dark zone of Alona and Panteleimonas caves,
Anthias anthias individual (B) photographed in the semi-dark zone of Alona and Panteleimonas caves,
respectively. Photos by T. Dailianis.
respectively. Photos by T. Dailianis.

All studied marine caves were significantly differentiated. In addition, one-way

ANOSIM analysis results indicated all examined geomorphological and topographic factors
(cave type, entrance depth, entrance area, and entrance orientation) as having significant
effects on the resemblance patterns of benthic community structure (p-value < 0.01 in all
cases; Table S2), as supported by the nMDS analysis shown on Figure 4. This was also
 All studied marine caves were significantly differentiated. In addition, one-way
ANOSIM analysis results indicated all examined geomorphological and topographic fac-
J. Mar. Sci. Eng. 2022, 10, 660 tors (cave type, entrance depth, entrance area, and entrance orientation) as having signif- 7 of 16
icant effects on the resemblance patterns of benthic community structure (p-value < 0.01
in all cases; Table S2), as supported by the nMDS analysis shown on Figure 4. This was
also supported
supported by thebypairwise
the pairwiseteststests (p-value
(p-value < 0.01< 0.01
in allincases)
all cases)
(Tables(Tables
S3–S5)S3–S5) verifying
verifying that
that the examined
the examined geomorphological
geomorphological andand topographic
topographic factors
factors significantly
significantly affect
affect thethe ben-
benthic
thic community
community patterns.
patterns. This This is particularly
is particularly evident
evident for theforsamples
the samples
from from different
different ‘eco-
‘ecological
zones’ (Figure
logical 4B), ‘cave
zones’ (Figure 4B),types’
‘cave(Figure 4C), and
types’ (Figure ‘entrance
4C), depth’ (Figure
and ‘entrance 4E) which
depth’ (Figure 4E)appear
which
to be grouped
appear in different
to be grouped clusters
in different in the in
clusters nMDS plots (i.e.,
the nMDS plotscave
(i.e., entrance vs. semi-dark
cave entrance vs. semi-
zone, semi-submerged vs. submerged caves and shallower vs. deeper
dark zone, semi-submerged vs. submerged caves and shallower vs. deeper caves, respec- caves, respectively).
According
tively). to the results
According to theofresults
SIMPER analysis,analysis,
of SIMPER the average the dissimilarity (based on(based
average dissimilarity the Bray-
on
Curtis
the dissimilarity
Bray-Curtis index) forindex)
dissimilarity different
for‘ecological zones’ waszones’
different ‘ecological 71.22%was while for ‘cave
71.22% types’
while for
and ‘entrance
’cave types‘ and depth’ was depth’
‘entrance 68.70%was and68.70%
69.82%, andrespectively. Among the
69.82%, respectively. studied
Among the caves,
stud-
average
ied caves,dissimilarity ranged between
average dissimilarity ranged 53.82%
between(Giourious vs. Oxonisos
53.82% (Giourious vs.caves) and 76.50%
Oxonisos caves)
(Alona
and vs. Panteleimonas
76.50% caves).
(Alona vs. Panteleimonas caves).

Figure
Figure 4.
4.Non-metric
Non-metricmultidimensional
multidimensional scaling (nMDS)
scaling (nMDS)plots showing
plots the similarities
showing among
the similarities the ana-
among the
lysed photoquadrat
analysed samples
photoquadrat from all
samples caves.
from all Photoquadrats are coloured
caves. Photoquadrats by different
are coloured by factors. (A)factors.
different Cave;
(B)
(A) Ecological Zone: entrance
Cave; (B) Ecological Zone:(CE) or semi-dark
entrance (CE) or (SD); (C) Cave
semi-dark (SD);Type: semi-submerged
(C) Cave (1) or sub-
Type: semi-submerged
merged (2); (D) Entrance Area: 15–40 m2 (1), 110–230 m 2 (2), or 630 m2 (3); (E) Entrance Depth (max):
(1) or submerged (2); (D) Entrance Area: 15–40 m2 (1), 110–230 m2 (2), or 630 m2 (3); (E) Entrance
0–10 m (1) and 10–20 m (2); (F) Entrance Orientation: east (E), west (W), north (N), and south (S).
Depth (max): 0–10 m (1) and 10–20 m (2); (F) Entrance Orientation: east (E), west (W), north (N), and
south (S).

Regarding the factor ‘entrance area’, samples from the caves with wider entrance
(group 2: 110–230 m2 and 3: 630 m2 ) seem to be grouped together and differentiated from
those with narrower entrance (group 1: 15–40 m2 ) in the nMDS plot (Figure 4D). SIMPER
 Regarding the factor ‘entrance area’, samples from the caves with wider entrance
(group 2: 110–230 m2 and 3: 630 m2) seem to be grouped together and differentiated from
those with narrower entrance (group 1: 15–40 m2) in the nMDS plot (Figure 4D). SIMPER
analysis showed lower dissimilarity between groups 2 and 3 (61.88%) than between 1 and
J. Mar. Sci. Eng. 2022, 10, 660 2 or 1 and 3 (66.43% and 73.18%, respectively). 8 of 16
According to SIMPER analysis for the factor ‘ecological zone’, the average similarity
of equivalent ’ecological zones‘ from different caves is higher (46.01% for CE and 37.16%
for SD) than
analysis showed the lower
average similarity between
dissimilarity between groups
the different
2 and 3zones (28.78%)
(61.88%) verifying1 that
than between and
equivalent ecological zones from
2 or 1 and 3 (66.43% and 73.18%, respectively). different caves show high similarity levels. The taxa
which contributed
According the most
to SIMPER to the
analysis for average
the factorsimilarity
‘ecologicalbetween
zone’, thephotoquadrats
average similarity wereof
Peyssonnelia sp., Bryozoan turf, and Encrusting Rhodophyta for
equivalent ’ecological zones‘ from different caves is higher (46.01% for CE and 37.16% forthe entrance zone (57%),
and Dendroxea
SD) than lenis, Serpulidae,
the average Bryozoan
similarity between theturf and Encrusting
different Bryozoa
zones (28.78%) for thethat
verifying semi-dark
equiva-
zone (53%) (Table
lent ecological zonesS6).from
Different taxa caves
different showed showdifferent
higher percent contribution
similarity levels. The to the
taxaaverage
which
similarity for every studied factor (Tables S7–S11).
contributed the most to the average similarity between photoquadrats were Peyssonnelia sp.,
Bryozoan turf, and Encrusting Rhodophyta for the entrance zone (57%), and Dendroxea lenis,
3.2. Threats and
Serpulidae, Pressures
Bryozoan turfonand
theEncrusting
Sessile Community
Bryozoa for the semi-dark zone (53%) (Table S6).
Different
Partialtaxa showed
necrosis wasdifferent
observed percent contribution
for different to theRhodophyte
sessile taxa. average similarity
necrosisforwasevery
rec-
studiedatfactor
orded (Tables of
the entrance S7–S11).
Oxonisos and Troulakas caves (0.8% and 0.5%, respectively) and
both entrance and semi-dark zones of Giourious (0.9% CE, 2.1% SD) and Panteleimonas
3.2. Threats
caves (2.9%and CE,Pressures
0.1% SD).onNecrosis
the Sessileof Community
the sponges Agelas oroides and Spirastrella cunctatrix
werePartial
recorded necrosis was observed
in the entrance for different
and semi-dark zonesessile
of the taxa. Rhodophyte
cave Oxonisos, withnecrosis was
0.2% cover-
recorded
age each.at the entrance
Partial necrosis ofofOxonisos
the sponge and Agelas
Troulakas caves
oroides and (0.8% andoros
Ircinia 0.5%,wererespectively)
also recordedand
both
at theentrance
entranceand semi-dark zones
of Panteleimonas, of Giourious
Giourious (Figure(0.9%
5A),CE,and2.1% SD) and
Oxonisos Panteleimonas
caves (Figure 5B).
caves (2.9%
Broken CE, 0.1%
fragments SD).coral
of the Necrosis pruvotiAgelas
of the sponges
Leptopsammia oroides
and erect and Spirastrella
bryozoans cunctatrix
(Myriapora trun-
were recorded in the entrance and semi-dark zone of the cave Oxonisos,
cata and Reteporella sp.) were also observed on the floor of the semi-dark zone of Troulakas with 0.2% coverage
each.(Figure
cave Partial necrosis
5C). of the sponge Agelas oroides and Ircinia oros were also recorded at the
entrance of Panteleimonas,
Another pressure observed Giourious
in most (Figure
of the5A), and Oxonisos
studied caves (Alona,caves (Figure 5B).
Oxonisos, Broken
Giourious,
fragments
and Troulakas)of the wascoral Leptopsammia
litter, mainly consistingpruvotiofand erect
plastic bryozoans
waste such as(Myriapora
fishing lines truncata
(Figureand5D
Reteporella sp.) were also observed on the floor of the semi-dark
and authors’ personal observations). Higher concentrations were observed in Oxonisos zone of Troulakas cave
(Figure 5C).
cave, a shallow semi-submerged cave, exposed to north winds and intense wave action.

Figure
Figure 5.
5. Partial
Partial necrosis
necrosis of
of the
the sponges
sponges Agelas
Agelas oroides
oroides (A)
(A) and
and Ircinia
Ircinia oros
oros (B)
(B) from
from the
the entrance
entrance zone
zone
of Panteleimonas and Oxonisos caves, respectively; fragments of the coral Leptopsammia pruvoti and
of Panteleimonas and Oxonisos caves, respectively; fragments of the coral Leptopsammia pruvoti and
the erect bryozoans Myriapora truncata and Reteporella sp. (C); fishing line (D) attached to the walls
the erect bryozoans Myriapora truncata and Reteporella sp. (C); fishing line (D) attached to the walls at
at the semi-dark zone of Troulakas cave. Photos by T. Dailianis.
the semi-dark zone of Troulakas cave. Photos by T. Dailianis.

Another pressure observed in most of the studied caves (Alona, Oxonisos, Giourious,
and Troulakas) was litter, mainly consisting of plastic waste such as fishing lines (Figure 5D
and authors’ personal observations). Higher concentrations were observed in Oxonisos
cave, a semi-submerged cave, exposed to north winds and intense wave action.
J.
J. Mar.
Mar. Sci.
Sci. Eng.
Eng. 2022,
2022, 9,
10,x 660
FOR PEER REVIEW 99 of
of 16
16

3.3. Motile
Motile Fauna
Fauna
In total, 45 motile taxa classified
classified as Pisces (24), Crustacea
Crustacea (10),(10), Echinodermata
Echinodermata (6), (6),
Polychaeta
Polychaeta (2),(2), Mollusca
Mollusca (2),
(2), and
and Mammalia
Mammalia (1) (1) were
were identified
identified as as species
species (41),
(41), genus
genus (2),
(2),
and family
family (2)
(2)ranks
ranks(Table
(TableS12).
S12).Among
Among thethe recorded
recorded motile
motile species,
species, fivefive are protected
are protected (Pal-
(Palinurus elephas,
inurus elephas, Scyllarides
Scyllarides latus,
latus, Epinephelus
Epinephelus marginatus,
marginatus, Paracentrotus
Paracentrotus lividus,
lividus, andand Mona-
Monachus
chus monachus)
monachus) under under the Bern
the Bern andand Barcelona
Barcelona Conventions
Conventions (Annex
(Annex II: List
II: List of endangered
of endangered or
threatened
or threatenedspecies and and
species Annex III: List
Annex III:ofList
species whose whose
of species exploitation is regulated).
exploitation Schools
is regulated).
of the circalittoral
Schools fish Anthias
of the circalittoral fishanthias
Anthias were visually
anthias wererecorded
visually at the entrance
recorded at theand semi-dark
entrance and
zone of Oxonisos
semi-dark zone ofand Panteleimonas
Oxonisos caves (Figure
and Panteleimonas 3B).(Figure 3B).
caves
In addition, 10 10 NIS were identified (Table S12). Non-indigenousNon-indigenous species species were
were re-
re-
ported at all studied cavescaves varying
varying fromfrom oneone for
for Giourious
Giourious cavecave toto seven
seven for
for Troulakas
Troulakas
cave (Table S12). Most
cave Most ofof the
the NIS
NIS were
were detected
detected only
only atat the entrance
entrance zone zone of the caves
forsskali, Siganus luridus, S. rivulatus, Torquigener flavimaculosus,
(Parupeneus forsskali, Siganus luridus, S. rivulatus, Torquigener flavimaculosus, and Diadema
(Parupeneus Diadema
setosum) (Figure 6A,B) while a few others were exclusively found
setosum) (Figure 6A,B) while a few others were exclusively found at the semi-dark partsat the semi-dark parts
(Urocaridella pulchella and Cerithium scabridum) or at both zones (Sargocentron
(Urocaridella pulchella and Cerithium scabridum) or at both zones (Sargocentron rubrum, rubrum, Pem-
pheris rhomboidea,
Pempheris rhomboidea, Pterois
and and miles)
Pterois (Figure
miles) 6C,D).
(Figure 6C,D).

Figure
Figure 6.
6. The
TheNIS
NISfish Siganus
fish luridus
Siganus (A)(A)
luridus andand
the the
echinoderm Diadema
echinoderm setosum
Diadema (B) from
setosum the entrance
(B) from the en-
zone of Troulakas and Palatia caves, respectively. The non-indigenous fish species Pempheris rhom-
trance zone of Troulakas and Palatia caves, respectively. The non-indigenous fish species Pempheris
boidea (C) and Pterois miles (D) from the semi-dark zone of Panteleimonas and Oxonisos caves, re-
rhomboidea (C) and Pterois miles (D) from the semi-dark zone of Panteleimonas and Oxonisos caves,
spectively. Photos by T. Dailianis.
respectively. Photos by T. Dailianis.
3.4.
3.4. Ecological
Ecological Quality
Quality Assessment
Assessment
The
The ecological quality of
ecological quality of the
the studied
studied marine
marine caves
caves was
was assessed
assessed asas poor
poor or
ormoderate,
moderate,
with values of CavEBQI ranging from 2.7 (poor) for Palatia cave to
with values of CavEBQI ranging from 2.7 (poor) for Palatia cave to 5.2 (moderate) for Alona5.2 (moderate) for
Alona cave (Table 4). The highest values were calculated for the two
cave (Table 4). The highest values were calculated for the two fully submerged caves (i.e., fully submerged
caves
Alona (i.e., Alona andwhich
and Toulakas), Toulakas),
had a which
moderatehadquality
a moderate
status.quality status. The
The Confidence Confidence
Index (CIEBQI )
Index (CI ) was high in all cases (Table 4), since the overall assessment
was high in all cases (Table 4), since the overall assessment was based on recent data,
EBQI was based on
recent data, following the recommended methodology [22]. Most
following the recommended methodology [22]. Most marine caves showed extremely low marine caves showed
extremely
percent coverlowofpercent
‘Passive cover
filteroffeeders’
‘Passive filter
(i.e., feeders’ (i.e.,
scleractinian scleractinian
corals), corals), ranging
ranging between 0.5 and
between 0.5 and0–25%),
3.7% (category 3.7% (category
while they 0–25%), while they
were totally absent were
fromtotally
Palatiaabsent
cave.from Palatia‘Large
The group cave.
The group
active filter‘Large
feeders’active
(i.e., filter feeders’ (i.e.,
massive/erect massive/erect
sponges, sponges,
large bryozoans, large bryozoans,
mollusks, mol-
and ascidians)
lusks, and coverage
had mean ascidians)between
had mean 0.9 coverage
and 10.2%between
(category 0.9 and 10.2%
0–25%) (category
in each cave. The0–25%)
groupin‘Small
each
J. Mar. Sci. Eng. 2022, 10, 660 10 of 16

active filter feeders’ (i.e., encrusting sponges, serpulids, and brachiopods) had mean cover
between 28.3 and 55.7% (categories 25–50% and 50–75%, respectively). Volumetric stratifica-
tion was low at the caves Alona, Giourious, and Palatia (centimetric-sized organisms) and
moderate for Troulakas, Panteleimonas, and Oxonisos (decimetric-sized organisms) due to
the presence of the massive sponge Agelas oroides, a few hydrozoans and bioconstructions
formed by serpulids and bryozoans. As far as motile taxa are concerned, ‘Cave-dwelling
mysids’ were observed in low numbers (a few individuals) only at the semi-dark interior of
Alona and Troulakas caves (Figure S1). The highest numbers of ‘Characteristic carnivores’
were found in Alona cave (6 species) and ‘Associated carnivores’ in Panteleimonas cave
(10 species), while the highest number of ‘Detritus feeders and omnivores’ were found at
Troulakas cave (4 species) (Table S12).

Table 4. Results of ecological quality index CavEBQI and Confidence Index CIEBQI for the six studied
caves.

Cave CavEBQI Ecological Quality CIEBQI Confidence Index

Alona 5.2 Moderate 10 High
Giourious 3.7 Poor 10 High
Oxonisos 3.8 Poor 10 High
Palatia 2.7 Poor 10 High
Panteleimonas 3.5 Poor 10 High
Troulakas 4.5 Moderate 10 High

4. Discussion
Although the number of studies focusing on marine caves of the Eastern Mediter-
ranean Sea has increased over the last years, few studies quantitatively assess their benthic
communities [10,32,33,44]. Herein, a quantitative description of sessile benthic communities
and a catalogue of motile taxa (including non-indigenous species) recorded in six marine
caves in the North Karpathos and Saria Islands MPA (Greece, Eastern Mediterranean Sea)
is provided along with the first ecological quality assessment for this habitat in the Eastern
Mediterranean Sea. The correlation of six different geomorphological and topographic cave
factors (i.e., cave, ecological zone, cave type, entrance depth, entrance area, and entrance
orientation) with the sessile community structure was also investigated. All studied marine
caves were significantly differentiated verifying the individuality hypothesis of the current
study (hypothesis i). Cave-specific (micro)topography with associated environmental gra-
dients, modifications in larval or trophic supply as well as stochastic biological patchiness
(e.g., several cave-exclusive and rare taxa recorded only from a few or a single marine cave)
may lead to cave individuality [2,4,9,45,46].
Factors, such as ’cave type‘, ‘entrance depth’, ’entrance area‘, and ‘entrance orientation’,
were found to be significantly associated with the benthic community patterns, verifying the
hypothesis that the examined geomorphological and topographic factors significantly affect
the cave community structure (hypothesis ii). Geomorphological features represent a proxy
for other environmental features in caves, such as light availability or water chemistry [47].
For instance, the hydrodynamic regime and light exposure increase in semi-submerged
caves, caves with lower entrance depth and/or wider entrance [2]. Orientation with respect
to the cardinal points (i.e., aspect) and presence of multiple entrances can also affect the
cave community patterns creating a variation on the level of light penetration and inner
cave parts’ isolation [2,3].
Marine caves showed higher resemblance when compared within equivalent ‘eco-
logical zones’, mainly due to the dominance of rhodophytes and sponges in the entrance
zone and semi-dark zone respectively [2], verifying that equivalent ecological zones from
different caves show higher similarity levels (hypothesis iii). Taxa distribution patterns
are in agreement with previous studies on marine caves of the Eastern Mediterranean
Sea [10,33,48].
J. Mar. Sci. Eng. 2022, 10, 660 11 of 16

Although photographic methods provide advantages, such as the ability to collect

high numbers of informative samples to characterize sessile assemblages within the limited
time of diving-operated surveys, some limitations also occur [49,50]. For this reason, taxa
whose identification is uncertain through external morphology and is only feasible through
detailed morphological and/or molecular analysis (e.g., bryozoans, and rhodophytes) were
grouped in morpho-functional categories (e.g., Bryozoan turf, Encrusting Rhodophyta). As
a result, cave species richness is potentially underestimated at both ecological cave zones.
Future sample analysis and monitoring, as well as quantitative visual census of motile
species could complement and expand the biodiversity lists of the studied caves.
Most of the sessile species identified are commonly found in the interior of marine
caves and other marine habitats of Greece while 10 species were reported for the first time
as part of the marine cave fauna of the Aegean Sea [2,27]. These were the demosponge
Haliclona (Reniera) aquaeductus, the gastropod Cerithium scabridum, the bryozoans Caberea
boryi, Reptadeonella violacea and Patinella radiata, the echinoderm Diadema setosum, the
ascidian Pycnoclavella nana and the fishes Atherina sp., Parupeneus forsskali, and Torquigener
flavimaculosus, enlarging the existing list of taxa [27]. The protected calcarean sponge
Petrobiona massiliana was recorded for the second time in Aegean marine caves [20].
Different sponges of the class Homoscleromorpha, categorised as Plakinidae and
Plakina sp., were reported from the semi-dark zone of three caves (Table S1). Although
four new species of the genus Plakina have recently been described from marine caves
of Greece, they are still poorly known due to their small size and cryptic habit [51,52].
Therefore, targeted sample analysis and future studies on marine caves of the Aegean Sea
may expand the known biodiversity of plakinid sponge species. Although brachiopods
exhibit a particular preference for cryptic habitats they have rarely been studied in marine
caves of the Eastern Mediterranean. The brachiopod species found in this study are among
the most abundant in marine caves of the North Aegean Sea [53]. Two circalittoral or
deep-water species were recorded in the studied caves, confirming the affinity of marine
cave fauna with that of deeper waters [2]. The circalittoral bivalve Neopycnodonte cochlear,
which was observed in one studied cave (Table S1, Figure 3A), has been reported to form
thick encrustations on rocky walls in marine caves of Italy and Croatia [54,55]. The cardinal
fish Anthias anthias, which formed schools in two marine caves (Table S12, Figure 3B), is a
nocturnal fish, usually observed at depths exceeding 30 m [56].
Despite the fact that visual surveys for motile fauna did not follow a quantitative as-
sessment protocol and they tend to underestimate small-sized and highly cryptic fauna [57],
two cryptobenthic fish species were recorded in the study area (Microlipophrys nigriceps and
Tripterygion melanurum) (Table S12). Further research using standardised protocols for the
assessment of motile assemblages, including cryptobenthic taxa, is expected to increase the
motile diversity of Aegean marine caves [57].
Seven out of 10 Decapoda species (Dromia personata, Lysmata seticaudata, Palaemon
serratus, Palinurus elephas, Plesionika narval, Scyllarides latus, and Stenopus spinosus) identi-
fied from the studied caves are considered to be among the most common decapods in
Mediterranean Sea caves [58]. The abundance of Plesionika narval in three marine caves
(Alona, Giourious and Palatia) (Table S12 and authors’ personal observations) indicates
that the marine cave ecosystem can act as refuge for this species, which is highly fished in
parts of the Aegean Sea. The endangered Mediterranean monk seal Monachus monachus, the
population of which is well known and monitored in the area over the last years [59] was
among the 12 protected species observed in the studied caves. Other interesting features of
scientific interest and conservation concern are the small-sized serpulid bioconstructions
observed in Panteleimonas cave (Figure 3B), which resemble those discovered in dark caves
of other Greek islands and are relatively smaller compared to those from Apulia, Sicily and
Cyprus [60–63].
Several threats were recorded at all studied caves. A total of 10 NIS were observed.
All recorded NIS have Indo-Pacific origin, having reached the Eastern Mediterranean
basin through the Suez Canal [64]. Although in the relevant literature there are no data
J. Mar. Sci. Eng. 2022, 10, 660 12 of 16

that indicate any direct impacts of NIS on the diversity of marine caves, at least three
of the observed NIS (i.e., Pterois miles, Siganus luridus and S. rivulatus) are considered as
invasive with high impact on the native biodiversity of the Aegean Sea and the Eastern
Mediterranean, while the remaining NIS have low or unknown impact [28,65]. Only two
NIS were reported exclusively in the semi-dark cave zone (Table S3), suggesting that
this habitat type could be unfavourable, at least to a certain point, for impacts related to
opportunistic NIS [28]. However, the high abundance of the sweeper Pempheris rhomboidea
at two caves (Oxonisos and Panteleimonas) (Table S12 and authors’ personal observations)
could probably have a high negative impact due to its predating behaviour [28], but further
future investigation is needed. The cleaner shrimp Urocaridella pulchella was recently
reported for the first time in Greek waters from several marine caves with the earliest
confirmed record in 2018 in Crete [30]. The current study fills a distributional gap for this
non-indigenous palaemonid, representing the first sighting from Karpathos Island. The
presence of NIS at all studied caves indicates their possible establishment.
Partial necrosis was observed for different sessile taxa such as rhodophytes and
sponges (Agelas oroides and Spirastrella cunctatrix). Such necrosis may be related with
seasonal environmental fluctuations [66], or broader mortality events due to temperature
rise in the Mediterranean Sea as it has been shown in previous studies [67–69]. Marine
litter was also among the pressures observed in most of the studied caves. Until today, very
few studies have focused on pollution in Mediterranean marine caves indicating our gap
of knowledge on some threats that caves are continuously facing [2,70].
The assessment of ecological quality revealed poor ecological quality for most caves
(all semi-submerged). Shallow semi-submerged caves are exposed to a higher hydrody-
namic regime caused by wave action and are vulnerable to sea surface temperature rise,
anthropogenic pressures (e.g., pollution) and higher numbers of NIS [2,13,23,24,28] which
can affect their community structure and ecological quality. It should be noted that the
ecological quality index (CavEBQI) applied in this study was developed for marine caves of
the Western Mediterranean basin [22]. Therefore, it remains still to be explored if the above-
mentioned results are linked to the biogeographic heterogeneity of the Mediterranean Sea,
indicating rather the oligotrophic conditions of the eastern basin than a possible decline
in habitat quality [13,71]. For instance, a striking example of how this assessment can be
affected is the notable absence of decimetric-sized passive filter feeders (e.g., gorgonians)
from shallow marine caves in the Eastern Mediterranean Sea [2,27] which cause a reduction
in volumetric stratification and thus underestimation of the ecological quality. In addition,
the impact of NIS on the ecological quality of the marine cave ecosystem should be also con-
sidered and included in the assessment process [72]. Given the scarcity of risk assessments
for marine cave ecosystems globally [73] and the general lack of information on the relation-
ships between cave species [74], further trials and cross-calibration exercises are already
under way in order to establish monitoring schemes in the Eastern Mediterranean Sea.
In conclusion, this quantitative and qualitative study revealed the rich biodiversity of
six marine caves of Karpathos and Saria Islands MPA. The results of the study highlighted
the spatial heterogeneity of hard substrate community composition among caves with
different geomorphology and topography. This heterogeneity should be considered in
future conservation and management actions [13,71]. In addition, monitoring and fur-
ther research is needed to deepen scientific knowledge on eastern Mediterranean marine
cave ecosystems.

Supplementary Materials: The following supporting information can be downloaded at: https:
//www.mdpi.com/article/10.3390/jmse10050660/s1: Table S1. Summary table with the average
percentage of coverage (%) of all sessile taxa and morpho-functional groups recorded at the ecological
zones of the six studied caves. Table S2. Results of one-way ANOSIM for six geomorphological and
topographic factors of the studied marine caves. Table S3. p-values of one-way ANOSIM pairwise
tests for the factor ‘Cave’. Table S4. p-values of one-way ANOSIM pairwise tests for the factor
‘Entrance Area’. Table S5. p-values of one-way ANOSIM pairwise tests for the factor ‘Entrance
Orientation’. Table S6. Similarity percentage analysis (SIMPER), showing the contribution of sessile
J. Mar. Sci. Eng. 2022, 10, 660 13 of 16

taxa to the average similarity (%) in each ecological zone of the studied caves. Table S7. Similarity
percentage analysis (SIMPER), showing the contribution of sessile taxa to the average similarity (%)
in the six studied caves. Table S8. Similarity percentage analysis (SIMPER), showing the contribution
of sessile taxa to the average similarity (%) in each cave type. Table S9. Similarity percentage analysis
(SIMPER), showing the contribution of sessile taxa to the average similarity (%) in each cave entrance
area range. Table S10. Similarity percentage analysis (SIMPER), showing the contribution of sessile
taxa to the average similarity (%) in each entrance depth range. Table S11. Similarity percentage
analysis (SIMPER), showing the contribution of sessile taxa to the average similarity (%) in caves with
different entrance orientation. Table S12. Summary table of the presence of motile species recorded in
the caves through visual census. Figure S1. Spider-web graphics representing the ecosystem-based
ecological quality evaluation of each studied marine cave.
Author Contributions: Conceptualization, V.G. and T.D.; methodology, V.G. and M.D.; Sampling,
V.G. and T.D.; software, M.D.; validation, M.D. and V.G.; formal analysis, M.D. and V.G.; investigation,
M.D., V.G. and T.D.; resources, T.D. and C.A.; data curation, M.D., V.G., T.D. and C.A.; writing—
original draft preparation, M.D.; writing—review and editing, V.G., T.D. and C.A.; visualization,
M.D.; supervision, V.G. and C.A.; project administration, T.D.; funding acquisition, T.D. and C.A. All
authors have read and agreed to the published version of the manuscript.
Funding: This research is funded in the context of the projects: (a) “Centre for the study and sus-
tainable exploitation of Marine Biological Resources (CMBR)” (MIS 5002670), which is implemented
under the Action “Reinforcement of the Research and Innovation Infrastructure,” funded by the
Operational Programme “Competitiveness, Entrepreneurship, and Innovation” (NSRF 2014-2020)
and co-financed by Greece and the European Union (European Regional Development Fund) and
(b) “Management interventions to N. Karpathos and Saria Protected Area”, included in the “South
Aegean Operational Programme 2014–2020” and co-financed by EU and national resources.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Data is contained within the article or Supplementary Materials.
Acknowledgments: The authors would like to thank Giorgos Chatzigeorgiou, Ioulios Glampedakis,
and Emmanouela Vernadou for their valuable help during fieldwork and assistance in parts of the
analysis as well as Michail Ragkousis for the creation of QGIS maps. Special thanks are due to the
personnel of the Management Agency of the Dodecanese Protected Areas, Konstantinos Protopappas,
Giorgos Priaris, Ioanna Papaggeli, Georgia Piligotsi, and the Head of the Management Agency,
Chariklia Kargiolaki, for their valuable operational support.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ouerghi, A.; Gerovasileiou, V.; Bianchi, C.N. Mediterranean marine caves: A synthesis of current knowledge and the Mediter-
ranean Action Plan for the conservation of “dark habitats”. In Marine Caves of the Eastern Mediterranean Sea: Biodiversity, Threats
and Conservation; Publication no. 53; Öztürk, B., Ed.; Turkish Marine Research Foundation (TUDAV): Istanbul, Turkey, 2019;
pp. 1–13.
2. Gerovasileiou, V.; Bianchi, C. Mediterranean marine caves: A synthesis of current knowledge. Oceanogr. Mar. Biol. Annu. Rev.
2021, 59, 1–88. [CrossRef]
3. Riedl, R. Biologie der Meereshöhlen; Paul Parey: Hamburg, Germany, 1966; pp. 1–636.
4. Bussotti, S.; Terlizzi, A.; Fraschetti, S.; Belmonte, G.; Boero, F. Spatial and temporal variability of sessile benthos in shallow
Mediterranean marine caves. Mar. Ecol. Progr. Ser. 2006, 325, 109–119. [CrossRef]
5. Russ, K.; Rützler, K. Zur Kenntnis der Schwammfauna unterseeischer Höhlen. Pubbl. Stn. Zool. Napoli 1959, 30, 756–787.
6. Sarà, M. Zonazione dei Porifera in biotope litorali. Pubbl. Stn. Zool. Napoli 1962, 32, 44–57.
7. Pouliquen, L. Les spongiaires des grottes sous-marines de la région de Marseille—Ecologie et systematique. Téthys 1972, 3,
717–758.
8. Balduzzi, A.; Bianchi, C.N.; Boero, F.; Cattaneo-Vietti, R.; Pansini, M.; Sarà, M. The suspension-feeder communities of a
Mediterranean Sea cave. Sci. Mar. 1989, 53, 387–395.
9. Martí, R.; Uriz, M.J.; Ballesteros, E.; Turon, X. Benthic assemblages in two Mediterranean caves: Species diversity and coverage as
a function of abiotic parameters and geographic distance. J. Mar. Biol. Ass. UK 2004, 84, 557–572. [CrossRef]
J. Mar. Sci. Eng. 2022, 10, 660 14 of 16

10. Gerovasileiou, V.; Dimitriadis, C.; Arvanitidis, C.; Voultsiadou, E. Taxonomic and functional surrogates of sessile benthic diversity
in Mediterranean marine caves. PLoS ONE 2017, 12, e0183707. [CrossRef]
11. Harmelin, J.G.; Vacelet, J.; Vasseur, P. Les grottes sous-marines obscures: Un milieu extrême et un remarquable biotope refuge.
Téthys 1985, 11, 214–229.
12. Gerovasileiou, V.; Aguilar, R.; Marín, P. Guidelines for Inventorying and Monitoring of Dark Habitats in the Mediterranean Sea;
SPA/RAC-UNEP/MAP, OCEANA: Tunis, Tunisia, 2017; pp. 1–40.
13. Giakoumi, S.; Sini, M.; Gerovasileiou, V.; Mazor, T.; Beher, J.; Possingham, H.P.; Abdulla, A.; Çinar, M.E.; Dendrinos, P.; Gucu,
A.C.; et al. Ecoregion-based conservation planning in the Mediterranean: Dealing with large-scale heterogeneity. PLoS ONE 2013,
8, e76449. [CrossRef]
14. Sini, M.; Katsanevakis, S.; Koukourouvli, N.; Gerovasileiou, V.; Dailianis, T.; Buhl-Mortensen, L.; Damalas, D.; Dendrinos, P.;
Dimas, X.; Frantzis, A.; et al. Assembling ecological pieces to reconstruct the conservation puzzle of the Aegean Sea. Front. Mar.
Sci. 2017, 4, 347. [CrossRef]
15. Abdulla, A.; Gomei, M.; Maison, E.; Piante, C. Status of Marine Protected Areas in the Mediterranean Sea; International Union for
Conservation of Nature (IUCN): Malaga, Spain; World Wide Fund for Nature (WWF): Paris, France, 2008; pp. 1–152.
16. Petricioli, D.; Buzzacott, P.; Radolović, M.; Bakran-Petricioli, T.; Gerovasileiou, V. Visitation and conservation of marine caves.
In Proceedings of the International Symposium “Inside and outside the Mountain”, Custonaci, Sicily, Italy, 23–26 April 2015;
pp. 29–30.
17. Chevaldonné, P.; Lejeusne, C. Regional warming-induced species shift in north-west Mediterranean marine caves. Ecol. Lett.
2003, 6, 371–379. [CrossRef]
18. Parravicini, V.; Guidetti, P.; Morri, C.; Montefalcone, M.; Donato, M.; Bianchi, C.N. Consequences of sea water temperature
anomalies on a Mediterranean submarine cave ecosystem. Estuar. Coast. Shelf Sci. 2010, 86, 276–282. [CrossRef]
19. Costa, G.; Betti, F.; Nepote, E.; Cattaneo-Vietti, R.; Pansini, M.; Bavestrello, G.; Bertolino, M. Sponge community variations within
two semi-submerged caves of the Ligurian Sea (Mediterranean Sea) over a half-century time span. Eur. Zool. J. 2018, 85, 381–391.
[CrossRef]
20. Gerovasileiou, V.; Dailianis, T.; Sini, M.; del Mar Otero, M.; Numa, C.; Katsanevakis, S.; Voultsiadou, E. Assessing the regional
conservation status of sponges (Porifera): The case of the Aegean ecoregion. Mediterr. Mar. Sci. 2018, 19, 526–537. [CrossRef]
21. Sempere-Valverde, J.; Sabino Lorenzo, Á.; Espinosa, F.; Gerovasileiou, V.; Sánchez-Tocino, L.; Navarro-Barranco, C. Taxonomic
and morphological descriptors reveal high benthic temporal variability in a Mediterranean marine submerged cave over a decade.
Hydrobiologia 2019, 839, 177–194. [CrossRef]
22. Rastorgueff, P.A.; Bellan-Santini, D.; Bianchi, C.N.; Bussotti, S.; Chevaldonné, P.; Guidetti, P.; Harmelin, J.G.; Montefalcone, M.;
Morri, C.; Perez, T.; et al. An ecosystem-based approach to evaluate the ecological quality of Mediterranean undersea caves. Ecol.
Indic. 2015, 54, 137–152. [CrossRef]
23. Nepote, E.; Bianchi, C.N.; Morri, C.; Ferrari, M.; Montefalcone, M. Impact of a harbour construction on the benthic community of
two shallow marine caves. Mar. Pollut. Bull. 2017, 114, 35–45. [CrossRef]
24. Montefalcone, M.; De Falco, G.; Nepote, E.; Canessa, M.; Bertolino, M.; Bavestrello, G.; Morri, C.; Bianchi, C.N. Thirty year
ecosystem trajectories in a submerged marine cave under changing pressure regime. Mar. Environ. Res. 2018, 137, 98–110.
[CrossRef]
25. Di Franco, A.; Milazzo, M.; Baiata, P.; Tomasello, A.; Chemello, R. Scuba diver behaviour and its effects on the biota of a
Mediterranean marine protected area. Environ. Conserv. 2009, 36, 32–40. [CrossRef]
26. Guarnieri, G.; Terlizzi, A.; Bevilacqua, S.; Fraschetti, S. Increasing heterogeneity of sensitive assemblages as a consequence of
human impact in submarine caves. Mar. Biol. 2012, 159, 1155–1164. [CrossRef]
27. Gerovasileiou, V.; Chintiroglou, C.; Vafidis, D.; Koutsoubas, D.; Sini, M.; Dailianis, T.; Issaris, Y.; Akritopoulou, E.; Dimar-
chopoulou, D.; Voutsiadou, E. Census of biodiversity in marine caves of the eastern Mediterranean Sea. Mediterr. Mar. Sci. 2015,
16, 245–265. [CrossRef]
28. Gerovasileiou, V.; Voultsiadou, E.; Issaris, Y.; Zenetos, A. Alien biodiversity in Mediterranean marine caves. Mar. Ecol. 2016, 37,
239–256. [CrossRef]
29. Ragkousis, M.; Abdelali, N.; Azzurro, E.; Badreddine, A.; Bariche, M.; Bitar, G.; Crocetta, F.; Denitto, F.; Digenis, M.; El Zrelli, R.;
et al. New Alien Mediterranean Biodiversity Records (October 2020). Mediterr. Mar. Sci. 2020, 21, 631–652.
30. Digenis, M.; Ragkousis, M.; Vasileiadou, K.; Gerovasileiou, V.; Katsanevakis, S. New records of the Indo-Pacific shrimp Urocaridella
pulchella Yokes & Galil, 2006 from the Eastern Mediterranean Sea. Bioinvasions Rec. 2021, 10, 295–303. [CrossRef]
31. Dailianis, T.; Gerovasileiou, V.; Chatzigeorgiou, G.; Vernadou, E.; Glampedakis, I.; Arvanitidis, C.; Protopappas, K.; Piligotsi, G.;
Kargiolaki, H. An extended baseline survey of cave habitats within a marine protected area in the Aegean Sea. In Proceedings of
the 6th European Conference on Scientific Diving, Freiberg, Germany, 21–22 April 2021.
32. Gerovasileiou, V.; Voultsiadou, E. Sponge diversity gradients in marine caves of the eastern Mediterranean. J. Mar. Biol. Assoc.
UK 2016, 96, 407–416. [CrossRef]
33. Dimarchopoulou, D.; Gerovasileiou, V.; Voultsiadou, E. Spatial variability of sessile benthos in a semi-submerged marine cave of
a remote Aegean Island (eastern Mediterranean Sea). Reg. Stud. Mar. Sci. 2018, 17, 102–111. [CrossRef]
34. Pérès, J.M. Mediterranean Benthos. Oceanogr. Mar. Biol. Annu. Rev. 1967, 5, 449–533.
J. Mar. Sci. Eng. 2022, 10, 660 15 of 16

35. Trygonis, V.; Sini, M. photoQuad: A dedicated seabed image processing software, and a comparative error analysis of four
photoquadrat methods. J. Exp. Mar. Biol. Ecol. 2012, 424–425, 99–108. [CrossRef]
36. Bacchiocchi, F.; Airoldi, L. Distribution and dynamics of epibiota on hard structures for coastal protection. Estuar. Coast. Shelf Sci.
2003, 56, 1157–1166. [CrossRef]
37. Marzinelli, E.M.; Underwood, A.J.; Coleman, R.A. Modified habitats influence kelp epibiota via direct and indirect effects. PLoS
ONE 2011, 6, e21936. [CrossRef] [PubMed]
38. Ostalé-Valriberas, E.; Sempere-Valverde, J.; Coppa, S.; García-Gómez, J.C.; Espinosa, F. Creation of microhabitats (tidepools) in
ripraps with climax communities as a way to mitigate negative effects of artificial substrate on marine biodiversity. Ecol. Eng.
2018, 120, 522–531. [CrossRef]
39. Sedano, F.; Pavón-Paneque, A.; Navarro-Barranco, C.; Guerra-García, J.M.; Digenis, M.; Sempere-Valverde, J.; Espinosa, F. Coastal
armouring affects intertidal biodiversity across the Alboran Sea (Western Mediterranean Sea). Mar. Environ. Res. 2021, 171, 105475.
[CrossRef] [PubMed]
40. Clarke, K.R.; Gorley, R.N. PRIMER v6: User Manual/Tutorial; PRIMER-E: Plymouth, UK, 2006; pp. 1–192.
41. Clarke, K.; Warwick, R. A Further Biodiversity Index Applicable to Species Lists: Variation in Taxonomic Distinctness. Mar. Ecol.
Prog. Ser. 2001, 216, 265–278. [CrossRef]
42. Mead, A. Review of the development of multidimensional scaling methods. J. R. Stat. Soc. Ser. D 1992, 41, 27–39. [CrossRef]
43. Clarke, K.R. Non-parametric multivariate analyses of changes in community structure. Austral. J. Ecol. 1993, 18, 117–143.
[CrossRef]
44. Galani, A.; Dailianis, T.; Sini, M.; Katsanevakis, S.; Gerovasileiou, V. Characterization of benthic communities in marine caves
of the island-dominated Aegean Sea (Eastern Mediterranean). In Proceedings of the 2nd Mediterranean Symposium on the
Conservation of Dark Habitats, Antalya, Turkey, 16 January 2019; pp. 67–68.
45. Morri, C.; Bianchi, C.N. Zonazione biologica. In Grotte Marine: Cinquant’anni di Ricerca in Italia; Cicogna, F., Bianchi, C.N., Ferrari,
G., Forti, P., Eds.; Ministero dell’Ambiente e della Tutela del Territorio: Roma, Italy, 2003; pp. 257–265.
46. Radolović, M.; Bakran-Petricioli, T.; Petricioli, D.; Surić, M.; Perica, D. Biological response to geochemical and hydrological
processes in a shallow submarine cave. Mediterr. Mar. Sci. 2015, 16, 305–324. [CrossRef]
47. Lunghi, E.; Manenti, R.; Ficetola, G.F. Cave features, seasonality and subterranean distribution of non-obligate cave dwellers.
PeerJ 2017, 5, e3169. [CrossRef]
48. Rosso, A.; Gerovasileiou, V.; Sanfilippo, R.; Guido, A. Bryozoan assemblages from two submarine caves in the Aegean Sea
(Eastern Mediterranean). Mar. Biodiv. 2018, 49, 707–726. [CrossRef]
49. Bianchi, C.N.; Pronzato, R.; Cattaneo-Vietti, R.; Benedetti-Cecchi, L.; Morri, C.; Pansini, M.; Chemello, R.; Milazzo, M.; Fraschetti,
S.; Terlizzi, A.; et al. Mediterranean marine benthos: A manual of methods for its sampling and study. Hard bottoms. Biol. Mar.
Mediterr. 2004, 11, 185–215.
50. Zapata-Ramírez, P.A.; Scaradozzi, D.; Sorbi, L.; Palma, M.; Pantaleo, U.; Ponti, M.; Cerrano, C. Innovative study methods for the
Mediterranean coralligenous habitats. Adv. Oceanogr. Limnol. 2013, 4, 102–119. [CrossRef]
51. Lage, A.; Araujo, H.P.M.; Gerovasileiou, V.; Muricy, G. A new cave-dwelling species of Plakina (Porifera: Homoscleromorpha)
from Crete, Greece (South Aegean Sea). Zootaxa 2018, 4466, 39–48. [CrossRef] [PubMed]
52. Lage, A.; Gerovasileiou, V.; Voultsiadou, E.; Muricy, G. Taxonomy of Plakina (Porifera: Homoscleromorpha) from Aegean
submarine caves, with descriptions of three new species and new characters for the genus. Mar. Biodivers. 2019, 49, 727–747.
[CrossRef]
53. Bitner, M.A.; Gerovasileiou, V. Taxonomic composition and assemblage structure of brachiopods from two submarine caves in
the Aegean Sea, Eastern Mediterranean. Eur. Zool. J. 2021, 88, 316–327. [CrossRef]
54. Taviani, M.; Angeletti, L.; Campiani, E.; Ceregato, A.; Foglini, F.; Maselli, V.; Morsilli, M.; Parise, M.; Trincardi, F. Drowned karst
landscape offshore the Apulian margin (Southern Adriatic Sea, Italy). J. Cave Karst Stud. 2012, 74, 197–212. [CrossRef]
55. Novosel, M.; Bakran-Petricioli, T.; Požar-Domac, A.; Kružić, P.; Radić, I. The benthos of the northern part of the Velebit Channel
(Adriatic Sea, Croatia). Nat. Croat. 2002, 11, 387–409.
56. Froese, R.; Pauly, D. FishBase. Available online: www.fishbase.org (accessed on 11 April 2022).
57. Ragkousis, M.; Digenis, M.; Kovačić, M.; Katsanevakis, S.; Gerovasileiou, V. Rarely Reported Cryptobenthic Fish in Marine Caves
of the Eastern Mediterranean Sea. J. Mar. Sci. Eng. 2021, 9, 557. [CrossRef]
58. Bianchi, C.N.; Gerovasileiou, V.; Morri, C.; Froglia, C. Distribution and Ecology of Decapod Crustaceans in Mediterranean Marine
Caves: A Review. Diversity 2022, 14, 176. [CrossRef]
59. MOm. Status Report of the Mediterranean Monk Seal Populations in Kimolos-Polyaigos and Karpathos Saria Areas; Final Report to EC
LIFE-NATURE 2000 Project (LIFE00NAT/GR/7248); Hellenic Society for the Study and Protection of the Monk Seal (MOm):
Athens, Greece, 2005.
60. Sanfilippo, R.; Rosso, A.; Guido, A.; Mastandrea, A.; Russo, F.; Riding, R.; Taddei Ruggiero, E. Metazoan/microbial biostalactites
from present-day submarine caves in the Mediterranean Sea. Mar. Ecol. 2015, 36, 1277–1293. [CrossRef]
61. Guido, A.; Jimenez, C.; Achilleos, K.; Rosso, A.; Sanfilippo, R.; Hadjioannou, L.; Petrou, A.; Russo, F.; Mastandrea, A. Cryptic
serpulid-microbialite bioconstructions in the Kakoskali submarine cave (Cyprus, Eastern Mediterranean). Facies 2017, 63, 21.
[CrossRef]
J. Mar. Sci. Eng. 2022, 10, 660 16 of 16

62. Rosso, A.; Sanfilippo, R.; Guido, A.; Gerovasileiou, V.; Taddei Ruggiero, E.; Belmonte, G. Colonisers of the dark: Biostalactite-
associated metazoans from “lu Lampiùne” submarine cave (Apulia, Mediterranean Sea). Mar. Ecol. 2021, 42, e12634. [CrossRef]
63. Kazanidis, G.; Guido, A.; Rosso, A.; Sanfilippo, R.; Roberts, J.M.; Gerovasileiou, V. One on Top of the Other: Exploring the Habitat
Cascades Phenomenon in Iconic Biogenic Marine Habitats. Diversity 2022, 14, 290. [CrossRef]
64. Katsanevakis, S.; Zenetos, A.; Belchior, C.; Cardoso, A.C. Invading European Seas: Assessing pathways of introduction of marine
aliens. Ocean Coast. Manag. 2013, 76, 64–74. [CrossRef]
65. Katsanevakis, S.; Zenetos, A.; Corsini-Foka, M.; Tsiamis, K. Biological Invasions in the Aegean Sea: Temporal Trends, Pathways, and
Impacts; The Handbook of Environmental Chemistry; Springer: Berlin/Heidelberg, Germany, 2020. [CrossRef]
66. García-Gómez, J.C.; González, A.R.; Maestre, M.J.; Espinosa, F. Detect coastal disturbances and climate change effects in
coralligenous community through sentinel stations. PLoS ONE 2020, 15, e0231641. [CrossRef] [PubMed]
67. Di Camillo, C.G.; Cerrano, C. Mass Mortality Events in the NW Adriatic Sea: Phase Shift from Slow- to Fast-Growing Organisms.
PLoS ONE 2015, 10, e0126689. [CrossRef] [PubMed]
68. Idan, T.; Goren, L.; Shefer, S.; Ilan, M. Sponges in a changing climate: Survival of Agelas oroides in a warming Mediterranean Sea.
Front. Mar. Sci. 2020, 7, 1064. [CrossRef]
69. Garrabou, J.; Gómez-Gras, D.; Ledoux, J.B.; Linares, C.; Bensoussan, N.; López-Sendino, P.; Bazairi, H.; Espinosa, F.; Ramdani, M.;
Grimes, S.; et al. Collaborative Database to Track Mass Mortality Events in the Mediterranean Sea. Front. Mar. Sci. 2019, 6, 707.
[CrossRef]
70. Mačić, V.; Ðord̄ević, N.; Petović, S.; Malovrazić, N.; Bajković, M. Typology of marine litter in “Papuča” (Slipper) cave (Montenegro,
South Adriatic Sea). Stud. Mar. 2018, 31, 38–43. [CrossRef]
71. Gerovasileiou, V.; Voultsiadou, E. Marine Caves of the Mediterranean Sea: A Sponge Biodiversity Reservoir within a Biodiversity
Hotspot. PLoS ONE 2012, 7, e39873. [CrossRef]
72. Katsanevakis, S.; Tsirintanis, K.; Sini, M.; Gerovasileiou, V.; Koukourouvli, N. Aliens in the Aegean—A sea under siege (ALAS).
Res. Ideas Outcomes 2020, 6, e53057. [CrossRef]
73. Mammola, S.; Meierhofer, M.B.; Borges, P.A.V.; Colado, R.; Culver, D.C.; Deharveng, L.; Delić, T.; Di Lorenzo, T.; Dražina, T.;
Ferreira, R.L.; et al. Toward evidence-based conservation of subterranean ecosystems. Biol. Rev. 2022. [CrossRef] [PubMed]
74. Lunghi, E.; Manenti, R. Cave communities: From the surface border to the deep darkness. Diversity 2020, 12, 167. [CrossRef]