Leptogorgia sarmentosa (Anthozoa: Octocorallia) in NE Sardinia (Mediterranean Sea): distribution and growth patterns

Distribution and occurrence data of the sea fan Leptogorgia sarmentosa (Esper, 1791) in the Mediterranean Sea were updated thanks to new photographic surveys, highlighting the presence of the species off NE Sardinia and, in particular, in the Tavolara-Punta Coda Cavallo Marine Protected Area (TPCCMPA). The estimated average distance among colonies strongly suggests that they are not able to sexually reproduce and that they probably can be considered a sterile pseudo-population supported by inputs of larvae from the Italian mainland coast towards Sardinia. The putative conveyor belt for larval input to this area is a small permanent gyre (the North-Tyrrhenian Gyre) counterclockwise moving from the Italian to the Sardinian coasts. In TPCCMPA, the growth of one colony settled at 30 m depth was investigated across 7 years (2016–2022). The morphometric parameters indicated a linear development of the colony, in terms of branch development and 3D complexity estimated through the fractal dimension. The observed growth rates were not regular but were reached through a highly variable positive and negative oscillation. A significant correlation was obtained between growth rate and average Autumn and Winter water temperature, suggesting that somatic growth occurs in these periods. In the context of long-lasting elevated seawater temperatures and local damage linked to fishing activities, L. sarmentosa was confirmed to be a resilient species compared with other erect seascape components, like other gorgonians and sponges.


Introduction
Leptogorgia sarmentosa (Esper, 1791) is a fan-shaped gorgonian, which is coloured with intense hues of orange, magenta, yellow or white, growing mostly in one plane and ramifying in a dichotomous way (Weinberg 1976). It is a ubiquitous and characteristic component of benthic northwestern Mediterranean assemblages (Gori et al. 2011;Rossi and Gili 2009), commonly living on gravel in soft bottoms and on horizontal, sediment-covered substrates (Pérès 1967), in turbid water and strong currents (Rossi 1965;Weinberg 1980). It plays an important role as an ecosystem engineer, providing biomass and structural complexity and sustaining rich biodiversity in benthic communities (True 1970;Gili and Ros 1985;Gili et al. 1989;Ballesteros 2006). The species usually occurs between 20 and 200 m depth (Carpine 1963;Weinberg 1976), although in some locations it lives in shallower water (Gori et al. 2011;Padrón et al. 2021) and, under peculiar conditions, reaching up the sea surface (Betti et al. 2018). It is a suspension feeder (Weinberg 1978;Gili and Ros 1985;Mistri 1995;Rossi and Gili 2009;Gori et al. 2011), with a higher growth rate compared to other octocorals (Mistri and Ceccherelli 1993), probably because of its high capture rates (Ribes et al. 2003;Rossi et al. 2004;Rossi and Gili 2009).
Thanks to recent surveys with the help of a remotely operated vehicle (ROV) and SCUBA coupled with citizen science reports, it is now possible to update the distribution of L. sarmentosa for the whole of the Mediterranean. From this analysis, it is evident that there is a gap regarding the Sardinian coasts, where the species appears to be absent.
The only exception is the area of the Marine Protected Area of Tavolara-Punta Coda Cavallo (TPCCMPA) off NE Sardinia. In this locality, thanks to photographic surveys, L. sarmentosa was already recorded along the limestone cliff of Tavolara Island and on granitic outcrops in the Tavolara Channel (Canessa et al. 2020(Canessa et al. , 2021.
The present study aims to update the distribution and occurrence of the species in the Mediterranean, check its presence in some localities along the NE Sardinia coasts and describe the population at TPCCMPA, together with the associated fauna living on the colonies. At the same time, a description of the growth pattern and development of a single colony over a period of 7 years is provided.

Occurrence and distribution
The geographic distribution of Leptogorgia sarmentosa in the Mediterranean Sea was assessed using data from various sources (Table 1) and citizen science records available online from four platforms: (1) iNaturalist; (2) the Global Biodiversity Information Facility (GBIF); (3) the openaccess database of the Reef Check Mediterranean Underwater Coastal Environment Monitoring protocol; (4) the Base pour l'inventaire des observations subaquatiques (BioObs). Data were managed in the QGIS 3.22 platform. At NE Sardinia, the species distribution was checked during 221 diving sites (two at La Maddalena, three at Santa Teresa di Gallura, 50 at Figari Cape, 166 within the Tavolara MPA) at depths of 5-65 m. Due to the protection policy of the area, the coordinates of the investigated sites are only available upon specific request from the park management. All the recorded colonies of L. sarmentosa were photographed. The multi-zoom photographic approach (Pititto et al. 2014) was used to characterize the geomorphology of the sites, the benthic assemblages, and in particular, the presence of L. sarmentosa, the occurrence of epibionts and putative damages due to human activities.

Growth pattern analysis
The morphology and growth pattern of a single colony of L. sarmentosa were studied at a dive site named "Occhio di Dio", at 30 m depth. This locality is characterized by a coralligenous assemblage and was monitored for 7 years (2016)(2017)(2018)(2019)(2020)(2021)(2022) with the help of images present in the collection of one of us (ET). Photographs were taken each year in May, using the same perspective and a ruler as a size reference. A Sony A6000 camera (24 megapixels, two Inon S2000 strobes, colour temperature 5000 K) with Sony 16-50 lens (focal length 19 mm), Nauticam WW1 wet wide lens (130°rectilinear field angle), and a Sea & Sea MDXA6000 underwater case with a flat porthole was used. For each image, height, width, basal diameter, number of primary branches, number of nodes and fan surface, as the total area occupied by the branches joining all the apexes, were measured and/or counted by ImageJ software (Rasband 2012). Moreover, the fractal dimension (FD) was evaluated each year according to the "box-counting method", randomly placing photographs of the gorgonian under six 256mm side grids partitioned into squares along each edge, with n varying from 2 to 7. For each image, the log-log plot of the number of squares entered by the outline of the gorgonian against the number of squares along one side of the grid was obtained through linear regression analysis: the slope of the resulting line equals the FD (Morse et al. 1985;Burlando et al. 1991).
For the total development of the branches, the gorgonian was isolated from the background, creating a binary image that was skeletonized, through a repeatable removal of pixels from the edges of objects until it is reduced to a single-pixel-wide shape. Finally, the total development resulted in the sum of pixels. To evaluate the putative impact of water temperature on L. sarmentosa growth, obtained data on the total development of the branches were correlated with the average seasonal SST of the years immediately before data collection (Summer: June-August; Autumn: September-November; Winter: December-February; Spring: March-May). SST data were derived from NOAA (US National Oceanic and Atmospheric Administration) satellite records, available at www.esrl. noaa.gov/psd/cgi-bin/data/timeseries/timeseries1.pl.

Distribution and occurrence
The occurrence of Leptogorgia sarmentosa across the Mediterranean basin indicates that the species is widely distributed in the Alboran Sea, the Sicily Channel, along the Italian Tyrrhenian coast, the Ligurian Sea and the Gulf of Lion, while sporadic occurrences are available for the Balearic Sea. Moreover, it is absent in the Ionian Sea, excluding records around Gallipoli. In the Adriatic Sea, the species is present on both sides, reaching the Istrian peninsula. Few records are also available for the Aegean Sea. Moreover, recent observations during the summer 2022 highlighted the presence of this species around the whole Corse (Fig. 1a).
The Capo Figari sites (Fig. 2c) were formed exclusively by landslides at the base of the limestone cliff and therefore similar to the sites of the Tavolara cliff and shoals. At La Maddalena and Santa Teresa di Gallura (Fig. 2d), the investigated sites were characterized only by granite reliefs and therefore can be compared with the granite outcrops of the Tavolara Channel. On September 24th, 2021, on the branches of one colony of L. sarmentosa, numerous spherical structures, putatively interpreted as eggs, were found (Fig. 2e, f).

Morphometric description and growth
The morphometric analysis of one colony of L. sarmentosa across 7 years allowed the description of its growth pattern over a long period of time (Table 2; Fig. 4). In general, the fan surface linearly increased (r = 0.714) from about 800 to 1800 cm 2 7 years later (Fig. 5a). Across 7 years, colony height almost doubled, from 31.47 to 55.71 cm (r = 0.85) (Fig. 5b). Similarly, the total length of the branches increased (r = 0.94), starting from about 6 m and eventually reaching about 21 m (Fig. 5c). The growth of these parameters was not regular, alternating years of positive and negative growth. A very similar linear trend (r = 0.99) was shown by the increase in the number of primary branches and nodes. In both these cases, the annual growths were very regular without periods of decrease (Fig. 5d-e). To investigate potential variation in the branching pattern complexity, the FD was estimated each year. For the studied specimen, FD increased from 1.3 to 1.7, with a linear positive trend (r = 0.88) (Fig. 5f).

Distribution pattern of Leptogorgia sarmentosa at Sardinia
The distribution of gorgonians is strongly constrained by the low dispersal ability of their larvae (Theodor 1967;Weinberg and Weinberg 1979) resulting in the philopatric distribution of their recruits (Gori et al. 2011). Padrón et al. (2018) stated that the primary filter of connectivity among gorgonian populations is larval transport, which is driven by the speed/ directionality of ocean currents and the time span over which larvae can be transported.
The present study of the gorgonian distribution in the Tyrrhenian Sea is a good opportunity to test these theoretical statements. In particular, Leptogorgia sarmentosa appears to be completely absent along the Sardinian coast with the only exception of rare records along the NE side. This area is hosting also unique Sardinian populations of the gorgonian Eunicella verrucosa ) and almost all known colonies of the zoantharian Savalia savaglia (Bertoloni, 1819) (Pulido Mantas et al. 2022). This unusual distribution pattern is validated by the fact that these species are generally well recognized by SCUBA divers and that they were never observed during the numerous ROV surveys off Sardinia Cau et al. 2015;Gori et al. 2019;Moccia et al. 2014Moccia et al. , 2021.
At the 221 diving sites of the present study, only 19 colonies of L. sarmentosa were found. Considering that the average distance between these colonies is about 870 m (Fig. 2b), probably that they do not derive from sexual reproduction in loco and are probably to be considered as a pseudo-population supported by inputs of larvae from other localities (Astraldi et al. 1995).
In the Mediterranean context, the density of L. sarmentosa is highly variable over different localities. In some cases, very dense forests were described (Weinberg and Weinberg 1979;Mistri 1995) generally associated with gravel bottoms (Pérès 1967) and strong water currents (Rossi 1965;Weinberg 1980). Nevertheless, in several cases, scattered specimens were recorded both on soft-bottom and coralligenous substrates (Carpine and Grasshoff 1975;Weinberg 1976;Gili et al. 1989;Mistri and Ceccherelli 1993;Rossi and Gili 2009;Gori et al. 2011). This kind of distribution, relatively unusual for Mediterranean shallow-water gorgonians, is probably due to the high survivorship of larvae in the plankton that can reach habitats far from the original population (Wangensteen et al. 2017). Under the effect of constant, unidirectional currents, sterile pseudo-populations may be established and maintained by the larval supply coming from the populations located upstream of the currents.
The gorgonian assemblage of NE Sardinia is very likely related to an input of larvae coming from the Italian mainland. The putative conveyor belt for larval input to this area is a small permanent gyre (the North-Tyrrhenian Gyre) counterclockwise moving from the Italian coasts to Corsica and Sardinia (Artale et al. 1994). The closest site along the Italian continental coasts is the Tuscany Archipelago (200 km) and this distance seems to be adequate to support this hypothesis. Although no data are available for the larval survival of L. sarmentosa, experimental tests on other Mediterranean gorgonians showed an average larval survive of 32-35 days very similar among different species (Guizien et al. 2020). At the same time, the Lion Gyre connects the Ligurian and the French coasts interested by the Ligurian-Provencal Current with the western part of Corse (Khélifi 2010). Additional genetic studies will be necessary to definitively clarify this complex scenario. The role of gyres in producing unexpected patterns of distribution in benthic organisms was already recognized. For example, the genetic population differentiation of the small benthic blennid Tripterygion tripteronotum Risso, 1810 corresponded with boundaries between the Adriatic re-circulation gyres (Sefc et al. 2020). In other cases, the gyre direction, coupled with the lack of suitable substrate, can constitute a barrier for the distribution of benthic organisms, like the threshold between the Tyrrhenian and the Ionian fauna (Villamor et al. 2014;Toma et al. 2022). Very likely, this barrier also prevents the colonization of the Ionian Sea by L. sarmentosa.
Sexual reproduction in L. sarmentosa was studied in detail by Rossi and Gili (2009), stating that spawning occurred between late August and early September, although there was no Fig. 4 a-g The binary images obtained for the investigated Leptogorgia sarmentosa colony, from 2016 to 2022; h Through the entire observation period, the identification of nodes (red dots) and I order branches (green triangles) allowed the reconstruction of the somatic colony development direct evidence. The present observations suggest a surface brooding strategy for this species. Surface brooding is an effective mechanism to allow male spermatic products to reach female colonies, without egg dispersion and with a substantial extension of the period in which eggs can be reached by sperms (Lasker 2006). Additional observations will be necessary to definitively confirm this hypothesis.
The long-term trend of the growth pattern Several authors have spent much effort to studying the growth pattern of L. sarmentosa, estimating the morphometric parameters of some colonies (height, fan surface, biomass) by photographic methods in time frames of up to 27 months . Alternatively, growth models have obtained records of morphometric parameters in several colonies whose age could be estimated based on the number of growth rings (Mistri 1995;Mistri and Ceccherelli 1993). Our attempt represents the first in which a single colony was monitored for a long time span (7 years) using the image analysis method, allowing the accurate evaluation of the total branch development. Our data support the finding that periodic photographic sampling is the best way to measure modular organism growth/shrinkage (Coma et al. 1998;Garrabou 1999;Garrabou and Zabala 2001;Lasker et al. 2003;Teixidó et al. 2009).
Over 7 years, our results clearly indicate a linear colony growth, mainly by the increase in the number and length of the first-order branches (Fig. 5). The height of the colony increased from 31.4 to 55.7 (3.4 cm year −1 ), the total branch development from 628 to 2130 cm (214.5 cm year −1 ), and the number of primary branches from 29 to 162 (19 per year). Although caution is due to the analysis of a single specimen, the growth rate of the colony height seems comparable with the 2.4-2.85 cm year −1 reported in the literature (Weinberg and Weinberg 1979;Mistri and Ceccherelli 1993). In contrast, Fig. 5 Growth pattern of Leptogorgia sarmentosa over a period of 7 years. a fan surface; b colony height; c total development of the primary branches; d number of primary branches; e nodes, f fractal dimension (FD) the basal diameter in the present study showed a negligible increase within a time span of 7 years.
The observed growth rates were not regular but reached through a highly variable positive and negative oscillation. Rossi et al. (2011) suggested that a water thermal anomaly caused the net negative growth observed in L. sarmentosa. Following this hypothesis, we have compared the annual growth rate, obtained in May with the average water temperature of the previous seasons (Fig. 6). While no correlations were recorded for spring and summer (Fig. 6a, b), a strong correlation was observed with the temperature of autumn and winter (Fig. 6c, d).
There is much literature about the effect of warm anomalies on the health of Mediterranean gorgonians (e.g. Cerrano et al. 2000;Garrabou et al. 2009Garrabou et al. , 2019. Nevertheless, L. sarmentosa is known to be an extremely resistant species, able to thrive in harsh environmental conditions, such as those typical of harbours (Betti et al. 2018). The species is known for its tolerance to extremely high temperatures since it is the only gorgonian not affected during periods of mass mortalities in the Mediterranean Sea (Sará et al. 2003). This is confirmed by the observations of one of us (ET) concerning massive sponges and anthozoans close to our study area, which were affected by elevated temperatures in the period of our observations.
Our data suggest that L. sarmentosa is not damaged by a warmer summer period if it is followed by a cool autumn-winter period. During summer, the energy effort is probably focused on reproduction occurring between late August and early September , while the somatic growth occurs in autumn-winter but only under adequate low temperatures. This is supporting the finding that, during spring and winter, the quantity of prey is about twice that of summer and fall in order to support the higher investment of the species in secondary production (Rossi 2002;Rossi et al. 2004). The effect of water temperature is likely in synergy with other both environmental and biological factors (Grigg 1974). For example, Mistri and Ceccherelli (1993) demonstrated that percental growth decreases with the age of the colonies, while Rossi et al. (2011) suggest that the oscillation of colony growth seems to be related to trophic constraints.
The branch loss due to epibiosis and predation is, very likely, negligible . In particular, L. sarmentosa seems to be a shallow-water gorgonian that is not much prone to mechanical injuries. This must be attributed to the high flexibility of its branches, which minimizes the damage due to fishing impacts (Bavestrello et al. 1997). Regarding predation, Mistri and Ceccherelli (1993) already stated that ovulid gastropods living on several specimens do not seem to cause serious damage to their hosts. Among epibionts, the large, branched ophiuroid Astrospartus mediterraneus was frequently observed. This species is in continuous expansion in shallow waters of the Tyrrhenian Sea . The observation of the persistence of a specimen on a L. sarmentosa colony at a multi-annual scale raises interesting questions about the ecology of this charismatic species. Mistri and Ceccherelli (1993) estimated that the colony complexity by fractal dimension indicates that this parameter remains stable (1.55) in colonies older than 1.5-2.5 years. Burlando et al. (1991) stated that in L. sarmentosa there is a positive correlation between the degree of development and the tendency to assume Fig. 6 Relationship between average water surface temperature and colony growth, calculated as percentage difference from the previous year. Negative correlation resulted for b Autumn (September-November) and c Winter (December-February) trimesters, but not in a Spring (March-May) and d Summer (June-August) a fractal geometry. This suggests that the gorgonian growth mechanisms retain a self-similar design, which becomes evident only in species combining a large size with a high branch density. Our data showing a progressive increase of the fractal dimension with age seems to confirm this hypothesis.
In conclusion, data here reported confirm that L. sarmentosa, under similar conditions, responds to the ongoing global warming process differently than other alcyonaceans (P. clavata and E. cavolini), and massive sponges (Axinella polypoides Schmidt, 1862, Ircinia oros Schmidt, 1864), demonstrating a capacity to grow continuously over time and limit mechanical impacts. In the context of rapid change of submerged habitats, where the effects of warming on erected species are amplified by the impacts of direct anthropogenic origin, the species can be considered a winner, considering the present stage of knowledge.
Acknowledgements The authors would like to thank the Tavolara MPA management for the permission to consult the GIS platform and for the possibility of independently developing the underwater surveys. The authors thank the "Slow Dive" team (M. Agostini, M. Munaretto, L. Nieddu, M. Romor and M. Tsuruoka for their support during the diving activities in the Tavolara MPA; F. Pisciottu and M. Muzzu for data regarding Santa Teresa; R. Romor for data regarding Capo Figari). A special thanks goes to Daniele Pauletto of Area 11 Diving Center for data concerning La Maddalena Archipelago NP and Lavezzi. We particularly thank Prof. Michele Scardi for the useful suggestions for the evaluation. We thank the editors and anonymous reviewers for their careful reading of our manuscript and their many insightful comments and suggestions.
Funding Open access funding provided by Università degli Studi di Genova within the CRUI-CARE Agreement. This study was funded by CN_00000033.

Declarations
Conflict of interest The authors declare no competing interests.
Ethical approval No animal testing was performed during this study.
Sampling and field studies All necessary permits for sampling and observational field studies have been obtained by the authors from the competent authorities and are mentioned in the acknowledgements.
Data availability Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.
Author contribution ET conducted field observation and photographic sampling. GB is responsible for the supervision, project administration and fund acquisition. MC validated and analyzed data. All authors wrote, reviewed and approved the manuscript.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.