Introduction

The Anthropocene1,2,3 (but see4) represents a new era in which anthropogenic forcings have profoundly modified ecosystem functioning, to such an extent that they constitute the main forcing. The effects of global change are evident in the structure, dynamics, functioning and resilience of marine and terrestrial ecosystems5,6,7,8,9,10,11,12,13. The cumulative impacts of the increase in direct and indirect anthropogenic forcings have a particularly pronounced effect on coastal ecosystems (e.g. habitat destruction, eutrophication, contamination, overfishing, invasive species, global warming, extreme events and modification of trophic networks). The various stresses act over time (short-term and/or long-term disturbances) in an additive and synergistic way on species, populations, ecosystems and their capacity to provide ecosystem goods and services (e.g6,7,14,15,16). Moreover, climate change can enhance the impact of other stressors, such as biological invasions and overfishing, with more pronounced short-term consequences for marine biodiversity and ecosystem functioning17. Overall, the Mediterranean Sea is impacted by the full spectrum of anthropogenic stressors, with synergistic effects making often the associated pressures and impacts more pronounced12,18.

Seaweed marine forest are the dominant habitat in pristine environments along temperate rocky coasts19,20,21. They play a significant ecological role in the structuring and functioning of the ecosystem, providing habitats, food sources and nurseries for a diverse range of species and providing ecosystem services15,20,22,23,24. The seaweed marine forests, which are mainly composed of Laminariales and Fucales (Phaeophyceae), exhibit high primary production, thereby influencing reef habitats and participating in the sustenance of diverse trophic levels20. Additionally, it plays a role in the carbon cycle25,26,27.

Most of the seaweed marine forests around the world, are in more or less marked decline (e.g15,20,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45), and this trend is expected to increase in relation with the climate warming for some species46. Nevertheless, some Cystoseira sensu lato populations (including Ericaria, Gongolaria and Cystoseira genera47) have maintained a relatively healthy state depending on locations43,48,49,50. The seaweed regression affects not only species but also the ecosystems in which they participate, as well as their ability to provide ecosystem services15. Following destruction, these macrophytes typically require several years to several decades to recover once environmental conditions have returned to an average state similar to that which preceded the disturbance (low resilience), but recovery may not occur, as a result of a MSS (multiple stable states) regime shift33,34,51,52 (but see53).

In the Mediterranean Sea, among the shallow species, Gongolaria barbata (Stackhouse) Kuntze [synonym: Cystoseira barbata (Stackhouse) C. Agardh] is a several years living (seven years in the Black Sea54), non-caespitose, species growing in very restricted marine habitats: very shallow, very sheltered, and well-lit reef habitats, rock-pools, and coastal lagoons. Gongolaria barbata is a long-lived a brown and non-iridescent species, characterised by an erect non-cespitose species with the presence of aerocysts. The species has a small blackish discoid spike, with single smooth cylindrical truncate axis without tophules and a protruding apex (shaped like a polished stone axe) (for more details see55).

The species sheds in winter floating secondary branches with aerocysts and receptacles. This is thus one of the few species of the genus Cystoseira sensu lato (the Cystoseira complex) able to spread over long distances. Gongolaria barbata is a species that occurs in the upper sub-littoral zone of the Mediterranean Sea (0.5–1 m depth), in sheltered locations and shallow bays, as well as in rock pools and coastal lagoons. The species shape the substrate by creating a 3-D habitat and host an important associated faunal and algal community. Their habitat is the one directly impacted by the destruction of habitats and by a strong anthropogenic and herbivory pressure55,56.

The development of macroalgal forests in the Mediterranean Sea, especially for Gongolaria barbata, is largely concentrated in shallow, sheltered bays, which are particularly susceptible to degradation as a result of urban development with the destruction of the habitats for example. These habitats, which are recognised for their high ecological value, are classified as ‘large shallow inlets and bays, code 1160’ within the Natura 2000 framework and protected under the European Union (EU) Habitats Directive, but not protected in France; in some instances, the degradation is so severe that natural recovery is nearly impossible in the short (years) or medium term (decades), even when the causes of their collapse have been removed. The natural recovery of Cystoseira sensu lato forests is frequently unachievable, even in instances where isolated natural recolonisation has been documented45,53,57. Restoration initiatives are frequently employed as a means of reinstating past ecosystem functions58. The practice of ecological restoration has emerged as a critical measure for the management of heavily degraded ecosystems59,60. The Convention on Biological Diversity has set restoration targets61, and, in 2020, the United Nations declared the ‘United Nation Decade of ecosystem restoration 2021–2030’62. This was followed by the proposed EU Nature Restoration Law63. There are few globally successful examples of active restoration that are focused on the recovery of canopy-forming species and success always occurred on a very small areas of few dozen of m² (e.g64,65,66,67,68).

The present study aims to (i) establish the current distribution of G. barbata along the French Mediterranean coasts including Corsica; (ii) reconstruct the past distribution using historical data; (iii) analyze the long-term evolution of the distribution to assess the status of the taxon and identify any potential causes of decline; and (iv) finally identify the potential benefits of implementing ecological restoration.

Materials and methods

Study area

The Mediterranean coasts of France was considered in its entirety, from Cerbère (at the Spanish border) to Menton (at the Italian border), including Corsica. The area was divided into five administrative regions: French Catalonia and Languedoc, Western Provence, Eastern Provence, French Riviera, and Corsica (Fig. 1).

Fig. 1
figure 1

Studied sites. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image

Data collection

Historical data

Historical records of Gongolaria barbata were analysed from a variety of sources, including published articles, unpublished reports and herbarium vouchers. The majority of the printed documents are kept in the ‘MACROPHYTES’ research platform of the Mediterranean Institute of Oceanography (MIO, Aix-Marseille University).

Vouchers held in herbaria represent an invaluable source of data allowing the verification of specimen identification. Our survey encompassed over 4 000 vouchers of Cystoseira sensu lato and Sargassum specimen held at the following Institutes (Herbaria acronyms after Thiers69):

  • Aix-Marseille University - Mediterranean Institute of Oceanography (HCOM): Herbarium C.F. Boudouresque, Herbarium H. & P. Huvé, Herbarium Saint-Charles, Herbarium Enseignement, Herbarium T. Thibaut and Herbarium M. Verlaque;

  • Institut National de la Recherche Agronomique - Botanical garden of Villa Thuret, Antibes (VTA): Herbarium Villa Thuret;

  • Muséum Requien (AV): Herbarium Requien;

  • Muséum National d’Histoire Naturelle de Paris (PC): Cryptogamic Herbarium, General Herbarium, Herbarium Bory de Saint-Vincent, Herbarium Cosson, Herbarium Feldmann, Herbarium Lamouroux, Herbarium Magne, Herbarium Montagne, Herbarium Lamouroux, Herbarium Le Prieur, Herbarium R. Melin, Herbarium Sauvageau, Herbarium Schousboe and Herbarium Thuret;

  • Musée Océanographique de Monaco (not indexed): Herbarium Mouret and Herbarium Børgesen;

  • Nice-Sophia Antipolis University (not indexed): Herbarium Meinesz;

  • Muséum d’Histoire naturelle de Nice (not indexed): Herbarium Camous, Herbarium Algues Vertes ;

  • Muséum d’Histoire naturelle de Marseille (not indexed): Herbarium Algue ;

  • Muséum d’Histoire naturelle de Toulon (not indexed): Herbarium Mouret ;

  • University of Montpellier (MPU): General Herbarium, Herbarium Flahault, Herbarium N. & F. Halle and Herbarium Raphelis;

  • Herbarium du Harmas de J.H. Fabre (FABR);

  • Musée et Jardins Botaniques Cantonaux MJBC (LAU): Herbarium de Lausanne and Herbier Ducommun.

Gongolaria barbata was found in Herbaria as Cystoseira hoppei C. Agardh, C. barbata var. hoppei (C. Agardh) J. Agardh and C. barbata f. hoppei (C. Agardh) Woronichin, C. barbata (Stackhouse) C. Agardh, C. barbata f. aurantia (Kützing) Giaccone and C. barbata f. repens A.D. Zinova & Kalugina.

Field work

The current distribution of G. barbata was investigated by field surveys conducted between 2007 and 2016, along the entire French Mediterranean coastline (~ 3 000 km of shoreline measured on a map with a 1:2 500 scale). The suitable habitat (very shallow sheltered upper sublittoral rocky zone) of G. barbata have been previously identified and selected on aerial photographs (Google Earth®) in order to optimize the field surveys. The upper littoral zone was explored by snorkelling. All the G. barbata populations were geolocalised and recorded on 1:2 500 maps.

GIS analyses

Each location (past or present) of Gongolaria barbata along the French Mediterranean coast was geolocalised and the past and present distribution patterns were analyzed on a GIS (Geographical Information System) database (ArcGis10®).

Statistical analyses

The present status of Gongolaria barbata in France was examined. We assessed whether previously reported populations of G. barbata were still present or could no longer be found, which could be indicative of possible local extinction.

The disappearance of a population in the very shallow, sheltered upper sublittoral rocky zone or in lagoons may be attributed to a number of potential causes. The species is not sensitive to a single cause within a station; rather, it may be exposed to several, including habitat destruction, the presence of invasive species, herbivory, uprooting or collection (as part of university teaching or research), and changes in environmental quality (such as desalination, pollution, silting, abrasion, etc.). In lagoons, fishing and sporting activities such as windsurfing or kitesurfing can also result in the species being uprooting. Uprooting a Cystoseira sensu lato that is not clumping (not-caespitose) is fatal for the individual (Table 1).

The impact of environmental changes has not been assessed in this study, due to the absence of significant environmental changes on a large scale. This is evidenced by the stability of the ecological status of water bodies, as evaluated with the CARLIT indicator (based upon mid-littoral and shallow macroalgae species)70.

Table 1 List of putative causes of a possible decline according to the French Mediterranean region considered.
Full size table

In our analysis of each locality, the major putative causes of decline were pointed out and 4 causes have been selected: habitat destruction, overgrazing, the presence of invasive species such as Sargassum muticum and other and unknown causes. Consequently, we decide to analyse for each past and present population according to the current status (still present, disappeared or newly recorded) and the major putative causes of decline through a multivariate exploratory approach using a non-metric multi-dimensional scaling (nMDS)71. Similarity measure matrices were calculated from the initial data matrix containing for each study region the number of localities which are still present or have disappeared, and the major putative anthropic pressures without transformation. The selected similarity measure was the S17 Bray Curtis similarity71,72. The nMDS represents samples as points in low-dimensional space such that the relative distances apart of all points are in the same rank order as the relative similarities of the samples. Finally, correlations of the current status and the major putative causes of decline, with the 2-D ordination plot of samples, were plotted by displaying correlation vectors. Spearman correlation was used given its non-parametric properties.

Results

Except in the coastal lagoon, where the species was often associated with Cystoseira compressa (Esper) Gerloff & Nizamuddin, in the open sea and all the regions, G. barbata was frequently associated with Ericaria crinita (Duby) Molinari & Guiry and, in a few cases, with Cystoseira foeniculacea f. tenuiramosa A. Gómez Garreta, M.C. Barceló, M.A. Ribera & J. Rull Lluch, as well as G. sauvageauana (Hamel) Molinari & Guiry and G. montagnei var. tenuior (Ercegović) Molinari & Guiry.

French Catalonia and Languedoc

During the 19th and the 20th century, the species was reported 20 times along the French Catalonia coast and 50 times along the Languedoc coast (Table S1). During the 20th century, the species was only reported along the Languedoc coast (26 times). The first record dates back to 1817, with specimens collected at Port-Vendres (specimens held in the Flahault herbarium at MPU) and the latest report dates from 1982 at Banyuls (Fig. 2, Table S1). In French Catalonia, the entire historical localities are no longer occupied by Gongolaria barbata.

The putative causes of the disappearance can be attributed to two main factors: overgrazing (90%) and habitat destruction (10%) (Table 2; Fig. 2). It is also important to consider the significant impact of intentional collections for academic courses on the populations of the rockpools of Banyuls-sur-Mer and Collioure.

The first record of this species in Languedoc dates back to 1835, with specimens collected at Sète (as Cette) (specimens held in the Requien herbarium at AV) and the latest report on open sea dating back to 1999 at Agde73 (Fig. 2, Table S1).

In the Languedoc, irrespective of whether the location is in the open sea or coastal lagoons, 59% of historical localities have not been found again, 16% are still present in a historical locality and 25% are new records (Table 1). It is necessary to consider the two situations in order to gain a nuanced understanding of the results: 55% of the localities are in open sea and 45% of the localities are in coastal lagoons. A review of the historical localities along the coast of Languedoc (open sea) revealed that 100% of these sites are no longer occupied by G. barbata. In contrast to the open sea, in coastal lagoons, only 9% of historical localities have disappeared, 35% are still present and 56% are new records. This persistence in coastal lagoons is unexpected as some of these lagoons are characterized be harsh conditions, in particular anoxic crises (malaïgues) and the competition with introduced seaweeds, especially the invasive Sargassum muticum. 60% of the open sea disappearance events can be ascribed to habitat destruction and 40% to overgrazing (Table 2; Fig. 3).

Table 2 Change in historical localities of G. Barbata (still present or absent, % of localities) and the major putative causes of observed decline, along the Mediterranean coasts of France.
Full size table
Fig. 2
figure 2

Gongolaria barbata in French Catalonia and Languedoc. black circle = new site; black cross = extinct in a historical site. Dates and sites in boldface: Gongolaria barbata still present or is currently present. Lagoons filled with cross bars (XXX) = presence of introduced seaweeds, especially of the invasive Sargassum muticum. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image
Fig. 3
figure 3

Two-dimensional nMDS ordination plot and correlation vectors (Spearman) in each study region, according to the current status and the major putative causes of decline (habitat destruction, overgrazing by sea-urchins, presence of the invasive species, especially Sargassum muticum, or other or unknown putative cause) of each historical locality.

Full size image

Western Provence

During the 19th and the 20th century, the species was collected 31 times along the Western Provence coasts (Fig. 4, Table S1). During the 21th century, the species has not been collected nor reported.

With regard to localities in the open sea, the first record of Gongolaria barbata dates back to 1815 in Marseille (specimens held in the Herbarium General at MPU) and the latest report of the species dates back to 1963 at Marseille (specimens held in the Herbier Huvé at HCOM). With regard to localities in lagoons, the first record of the species dates back to 1862 in Étang de Berre (specimens held in the Herbarium Huvé at HCOM) and the latest report dates back to 1970 at Étang de Berre (Saint-Chamas)74 (Fig. 4, Table S1). The entire historical localities in both open sea and lagoons are no longer occupied by G. barbata. During the field survey, we were unsuccessful in sighting the species.

The putative causes of its disappearance are attributed to 74% by overgrazing and 26% by habitat destruction. (Table 2; Fig. 3). Additionally, anoxic crises and pollution in Étang de Berre can also be taken into consideration.

Fig. 4
figure 4

Gongolaria barbata in Western Provence. black cross = extinct in a historical site. Dates and sites in boldface: G. barbata still present or currently present. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image

Eastern Provence

During the 19th and the 20th century, the species was collected 17 times along the Eastern Provence coasts and 86 times during the 21th century (Fig. 5, Table S1).

The first record of the species in Eastern Provence dates back to 1818 at Toulon. The specimen has been found in the Herbier Requien held in Muséum Requien (AV). Before the 21th century, Gongolaria barbata has been collected mainly to the west part of the eastern Provence, between Saint-Cyr-sur-Mer and Toulon but in the 19th century, G. barbata was also collected at Fréjus (General Herbarium, MPU) (Fig. 5, Table S1).

Many new records (21th century) were reported during our field survey (Fig. 5, Table S1). In the region, 13% of historical locality have not been found again, 8% have been still present in a historical site and 80% are new records (Table 2). The putative causes of their disappearance in historical localities are due to 92% by habitat destruction (Fig. 3; Table 2). The species has been found in sites favourable for its development.

Fig. 5
figure 5

Gongolaria barbata in Eastern Provence. White circle = still present in a historical site; black circle = new site; black cross = extinct in an historical site. Dates and sites in boldface: G. barbata still present or currently present. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image

French Riviera

During the 19th and the 20th centuries, the species was collected 59 times along the French Riviera coastline and only 12 times during the 21th century (Fig. 6, Table S1).

The first record dates back to 1825, with specimens collected along the coast at Nice (specimens held in the Sauvageau Herbarium, PC). In the 19th century, G. barbata was collected at Cannes (Raphélis Herbarium, Raphélis75), at Antibes (Saint-Charles Herbarium, Huvé Herbarium HCOM, Sauvageau Herbarium PC, Flahault Herbarium MPU, Le Prieur Herbarium, Villa Thuret Herbarium, Bornet & Flahault76), at Nice (Sauvageau Herbarium, Montagne Herbarium PC, Algues Vertes Herbarium – Muséum Histoire Naturelle de Nice), and at Menton (Bonfils in Raphélis77) (Fig. 6, Table S1).

Fig. 6
figure 6

Gongolaria barbata in French Riviera. White circle = still present in a historical site; black circle = new site; black cross = extinct in an historical site. Dates and sites in boldface: G. barbata still present or currently present. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image

At the beginning of the 20th century, vouchers specimens and records were numerous and the species was collected at Saint-Jean-Cap-Ferrat (J. Feldmann Herbarium, Carnet de récolte Feldmann, Ollivier78), Beaulieu-sur-Mer78, Roquebrune Cap-Martin (Bonfils in Raphélis75), and Menton (Raphélis Herbarium, MPU). The species was reported to be relatively common at Théoule (Raphélis Herbarium), Cannes (Raphélis Herbarium; Raphélis75), Golfe-Juan75, Antibes78, Nice (Camous Herbarium) and between Saint-Jean-Cap-Ferrat and Eze-sur-Mer in 191279. Ollivier78 described the species as being particularly abundant in the shallow bays located in the eastern and the western regions of Antibes, but he noted that the species was rare on the steep coast from Nice to Menton. Finally, Guglielmi80 reported the species as common at Villefranche-sur-Mer, Saint-Jean-Cap-Ferrat, and Beaulieu-sur-Mer (Fig. 6, Table S1).

The species has become increasingly rare since the 1970s, with only a single citation81 and two specimens collected at Saint-Jean-Cap-Ferrat in the 1970s and Antibes in the 1980s (Meinesz Herbarium) (Fig. 6, Table S1).

The current status of G. barbata is critical. The species was not found in most of its historical sites. It survives at only 3 sites: Saint-Marguerite Island, Antibes and Saint-Jean-Cap-Ferrat. In the region, 76% of historical sites have not been found again and 24% are new records (Table 2). The putative causes of their disappearance are attributed to 80% by habitat destruction and 20% by overgrazing (Table 2). It is worthy of note that intentional collection of specimens for higher education can have a tangible impact within the extensive rock pools at Saint-Jean-Cap-Ferrat.

Corsica

During the 19th and the 20th century, the species was collected 15 times along the Corsica coasts and 29 times during the 21th century (Fig. 7, Table S1).

In Corsica, the first record dates back to the 19th century in Ajaccio and Calvi (Fig. 7, Table S1). Gongolaria barbata was collected in two lagoons during the 1970s in Étang de Diane and Étang d’Urbino (Casabianca et al.82; Herbier Verlaque HCOM). However, these two localities are no longer occupied by the species.

In coastal lagoons, all historical localities are no longer occupied by G. barbata, in contrast to the open sea (which represents 93% of the sites). The situation is markedly different in open sea: 27% of historical sites have not been found again, 12% are still present in a historical locality and 61% have been newly recorded (Table 2).

Among the historical sites that have been lost, the putative causes of their disappearance are as follows: 36% can be attributed to overgrazing, 36% to habitat destruction, 29% to unknown causes and 7% to the presence of the invasive species such as Sargassum muticum (Fig. 3; Table 2). This last putative cause is the one involved in the disappearance of the historic sites of the Corsican lagoons.

The species is found in sites favourable for its development in coastal rocky shore.

Fig. 7
figure 7

Gongolaria barbata in Corsica. White circle = still present in a historical site; black circle = new site; black cross = extinct in an historical site. Dates and sites in boldface: G. barbata still present or currently present. Lagoons filled with crosses (XXX) = presence of the invasive Sargassum muticum. The map was created by Aurélie Blanfuné using ESRI/ArcGis10® software.

Full size image

Discussion

The Mediterranean Fucales belong to perennial species that structure their habitats. Their tolerance to disturbances (ecological plasticity) varies according to the species83. For example, G. barbata displays a higher level of ecological plasticity than other species. In the Adriatic Sea, this species shows ecological plasticity and remarkable morphological variability due to environmental factors84,85,86. In the Black Sea, the ecological plasticity might depend on the hydrodynamic regime and season87. Moreover, the presence of genetically disconnected populations at small spatial scales (less than 10 km) suggests that each population of G. barbata should be considered as separated evolutionary units for conservation efforts88. Verdura et al.89 highlight the necessity of mapping and standardized monitoring of Cystoseira s.l. forests in order to obtain baseline information for future management strategies involving their conservation, the mitigation of the stressors threatening them and the restoration of the degraded forests at a regional scale. It is crucial not to limit action only to the local scale.

This present study can serve as a baseline for the future monitoring. Verdura et al.89 emphasized the necessity of establishing reliable baselines for the distribution, the ecological status and stressors which are currently threatening Cystoseira s.l. forests in the Mediterranean Sea.

For the first time, an exhaustive map of the distribution of Gongolaria barbata populations has been produced at a very local scale (2:2 500) for the entire French Mediterranean coast, including Corsica. Similar studies have been conducted at the same scale for Ericaria amentacea (C. Agardh) Molinari & Guiry48 and for Ericaria crinita (Duby) Molinari & Guiry44, and at the regional scale for several species of Fucales29,43,49,90. The current distribution of G. barbata has been evaluated in comparison with historical data obtained from a comprehensive analysis of all available literature, grey literature and herbarium vouchers since 1815.

Accordingly, the status of Gongolaria barbata can be evaluated in accordance with the IUCN Red List criteria91. This taxon has not yet been assessed for the IUCN Red List and has been categorized as ‘Not Evaluated’ (NE)92. In the light of the comparison between historical and the current records, as well as the knowledge of anthropogenic pressures at each locality, we propose the IUCN-like status of G. barbata for the five French Mediterranean regions.

In French Catalonia and Western Provence, G. barbata should be classified as ‘Regionally Extinct’ (RE). In Languedoc, G. barbata should be classified as ‘Vulnerable’ (VU), in accordance with the IUCN criterion VU A2ac. In the latter region, the species is extinct in the open sea and still present in some coastal lagoons despite the considerable competitive pressure exerted by introduced seaweeds, especially the invasive Sargassum muticum. In French Riviera, G. barbata should be classified as ‘Critically Endangered’ (CR) under the IUCN criterion CR A2ac: more than 75% of lost historical localities, the species being now functionally extinct (even if the species is still present, it no longer fulfils its former ecosystem role. Finally, in Eastern Provence and Corsica, G. barbata should be classified as ‘Least Concern’ (LC) under IUCN criteria91.

The conservation status of only a small number of algal species has been properly assessed, primarily due to a lack of historical data. The information provided here could be used as a basis for improving the evaluation of the conservation status of G. barbata, an ecologically significant species and a regionally protected marine alga93,94. This evaluation of status according to the IUCN Red List criteria has previously been carried out for just one other Cystoseira sensu lato taxon, namely Ericaria crinita (Duby) Molinari & Guiry along the French Mediterranean coast44 and for one Laminariales, namely Laminaria rodriguezii Bornet for the Adriatic Sea95.

The status of G. barbata populations and the putative causes of decline vary between the regions under consideration. For instance, in French Catalonia and Western Provence, the major putative causes are overgrazing, habitat destruction and uprooting whereas in Languedoc, Eastern Provence and French Riviera, habitat destruction is the main cause. In Corsica, overgrazing and habitat destruction are involved in the same proportions. Some Cystoseira s l. species such as G. barbata, have historically played a significant functional role in the Mediterranean sublittoral reef ecosystems96,97,98. On this basis, in French Riviera, the species was already considered as functionally extinct, the surviving population no longer playing a significant role in ecosystem functioning42; the same have been noticed in Menorca Island (Spain)99. A possible reason of stable population of Gongolaria barbata in Eastern Provence and Corsica in comparison to other region is a lower destruction rate of irreversible destruction of shallow bottoms by reclamations. For instance, the rate in Corsica is 0.84%, whereas the French Riviera exhibits a rate of 20.18%100 Despite in Eastern Provence the rate of irreversible destruction of shallow bottoms by reclamations is 9.95%, G. barbata thrives in all the sites that are favourable to its development, which corresponds to a substantial proportion of the rocky linear coast in this area. As many canopy forming seaweeds, Fucales are highly sensitive to grazing. A pressure of grazing possibly lower than in other regions may play a role in the resilience of G. barbata in Corsica and Eastern Provence. In these two regions, recreational sea urchin fishing by snorkelling is a cultural practice and is much more prevalent than in other regions. Furthermore, for example in Western Provence, professional fishing is undertaken by scuba diving, which deflects the harvesting pressure towards the depth, where Gongolaria barbata is rare or absent101. Unfortunately, accurate data on grazing pressure are missing in the literature but obvious in the field.

As all the Fucales, this species has a monogenetic life cycle with a dominant sporophyte phase, while the gametophyte is reduced to just a few cells within the sporophyte102. Unlike countless algae species where the sporophyte and gametophyte phases are separated into distinct individuals, the lack of distinct individuals makes these taxa particularly vulnerable. Indeed, when all individuals are destroyed (uprooted, eaten) where gametophytes disappear along with the sporophytes, there is no chance for population resilience, as the gametophytes cannot contribute to recruitment like in the case of others habitat-forming species as the kelp (Laminariales)103. Although the species can float and potentially disperse, this is insufficient to compensate for the loss of these isolated populations.

It is essential that the conservation measures for the species must take into account the temporal distribution of populations, the anthropogenic pressures present in the area and the ecosystem to which it belongs, in the framework of an ecosystem-based approach104. The health of shallow water Cystoseira s.l. species thriving in sheltered areas depends on a wide range of factors, including geomorphological features, wave exposure and anthropogenic pressures105. Consequently, the putative causes of the decline of G. barbata populations could be multiple and cumulative. In several areas of the Mediterranean Sea and also in Black Sea, the local populations of G. barbata have exhibited a decline. The potential contributing factors identified include the artificialisation of the coastline, a deterioration in water quality and an increase in sediment loads56,87,105,106,107. It has been documented that overgrazing, whether by sea urchins or fish, has been a significant factor contributing to the extensive decline and local extinctions of numerous Cystoseira s.l. and Sargassum species in the Mediterranean Sea29,108,109,110,111,112.

The restoration of key species is promoted on an international scale, with the objective of reversing the observed trends of decline in the marine forest, with the ultimate goal of re-establishing ecosystem services (Biodiversity Strategy to 2020 – European Commission, 2011; Biodiversity Strategy to 2030 – European Commission, 2020; UN Decade on Ecosystem Restoration (2021–2030) – General Assembly of the United Nations, 2019).

What measures can be taken to conserve Gongolaria barbata populations, where the decline is significant enough to require restoration? Natural recovery should be preferred, but the restricted dispersal capacity of Cystoseira s.l. zygotes generally hinder natural recovery21,49,65,113.

Gongolaria barbata, the climax of one of the habitats of the infralittoral rock ecosystem with photophilic algae, represents one of the two stable states with the barren-grounds. All the intermediate stages of succession including colonisation by turf are possible. This ecosystem operates under top-down control, i.e. it is controlled by predators (top predators and sea urchin predators) whose overexploitation leads to a proliferation of herbivores (sea urchins and Sarpa salpa) that eliminate the Fucales forests. The ongoing disappearance of these populations is a sign of a profound and irreversible change in the ecosystem, leading to a state of barren-ground. Fucales (including Gongolaria barbata) is an integral component of the marine ecosystem, contributing significantly to its biodiversity and productivity. It plays a crucial role in shaping the spatial dynamics of the environment and serves as a nursery for many teleost species. Furthermore, seaweed facilitates the transition of various organisms, such as sea urchins and fish, to subsequent stages of their life cycle, thereby contributing to the complexity and stability of the marine food web (e.g114,115,116,117,118,119,120,121,122). Consequently, natural or experimental restoration of extinct or ecologically more functional populations only makes sense if it is part of an integrated management of human activities, and if it covers a large area where pressures have been reduced to a level that the Fucales can tolerate.

One potential solution is the implementation of active ecological restoration. However, there are several prerequisites that must be met before such a process can be initiated. Firstly, the species of interest must have historically been present at the site of restoration. Secondly, the putative causes of their disappearance must be identified and no longer present.

The first attempt in France to restore G. barbata via transplantation was made in the French Riviera (France), with no control of major herbivores [the fish Sarpa salpa (L., 1758) and the sea urchin Paracentrotus lividus (Lamarck, 1816)]. This attempt failed123.

The distribution of G. barbata has drastically declined in the Italian part of the Gulf of Trieste and currently, the species is only present in Slovenia124. Consequently, numerous restoration attempts have been undertaken in this area (e.g65,86,107,124,125,126,127,128,129,130). The restoration of Fucales forests may be done via transplantation of adult thalli107,131,132, the use of bags with fertile receptacles in situ65,128, or from outplanting juveniles grown ex situ in laboratory65,124,126,127,129,131,133,134,135.

A survey of the re-introduction of G. barbata in Menorca (Spain) highlighted the restored population expanding its surface area by three orders of magnitude (from 3.6 m² to 293 m²) in 10 years99, despite the limited dispersal ability of Cystoseira s.l. zygotes21,65,113 (but see49). Gongolaria barbata demonstrates a remarkable ability to proliferate and expand in a restoration site99. The impact of herbivory pressure from fish and sea urchins represents a significant threat to the survival and growth of the Cystoseira s.l. species. The implementation of control measures, including caging, has been shown to enable the success of the restoration (e.g86,90,111,136,137,138). A substantial proportion (60%) of the gut contents of Sarpa salpa can be composed of Cystoseira139. This herbivore has the capacity to remove up to 90% of the surface area of transplanted adults within a few days111. Additionally, mesograzers [e.g. Clibanarius erythropus (Latreille, 1818), Idotea balthica (Pallas, 1772), and Cerithium vulgatum Bruguière, 1792) contribute also to the herbivory pressure and hinder the success of the recruitment of Cystoseira s.l. species while eating them140. Interestingly, there is a paradox in French regulation: while Paracentrotus lividus is increasingly protected from harvesting, and thousands of juvenile sea urchins are being released, EU directives simultaneously classify G. barbata and the seaweed marine forests as strictly protected. Addressing this paradox is crucial for enabling the regulation of sea urchin populations to mitigate the impact of herbivory. This policy-level intervention can help reduce the adverse effects of anthropogenic activities, thereby enhancing the prospects of successful ecological rehabilitation. Such an approach will require a thorough assessment of grazing pressure in the reintroduction areas, followed by targeted regulatory actions.

It is worth noting that extreme climatic events such as marine heatwaves can influence the fertile period of the species, its settlement and its survival. Gongolaria barbata is typically fertile during the spring season55 and after a marine heatwave, the species may exhibit premature fertility in the winter141. Marine heatwaves have also been shown to negatively impact the zygote settlement and survival of early stages in Ericaria crinita142.

Bottom-up forces are fundamental, and necessary to account for the responses of ecosystems to perturbations, but they are not sufficient. Therefore, top-down forcing must be included and considered for optimal management solution and the restoration of marine forest143. It is more cost-effective and straightforward to implement protective measures (passive restoration) than to undertake active restoration. It is imperative to devise and implement a regional strategy for regulating herbivores, as a matter of urgency, in order to safeguard the marine forests and the ecosystem services they provide86,124,144,145. An ecosystem-based approach, coupled with population genomics studies146 is required to effectively manage and restore properly the infralittoral vegetation and ecosystems, which necessitates the management of herbivores on a large scale, both in Marine Protected areas (MPAs) and outside MPAs.

Conclusion

The results of this study refer to the French Mediterranean coast but the procedures can be applied everywhere not only of regional scale for many other species facing similar consequences in different parts of the world to evaluate their current status. Once the conservation status of a species has been established, it is important to understand the causes of any decline. In the case of a drastic decline, the implementation of active ecological restoration with mitigation measurement may be a solution. However, we must not forget the prerequisites (the species of interest must have been historically present at the restoration site and the putative causes of their disappearance must be identified and no longer present). It is also important to understand how the ecosystem functions so that restoration is ecosystem-centred and not just species-centred.