Impacts of the invasive species Caulerpa cylindracea Sonder, 1845 on the algae flora of the west coast of Algeria

Abstract The assessment of the impacts of the expansion of the invasive species on taxonomic diversity, the abundance and dominance of groups of algae, the presence and/or absence of species of ecological interest that may or may not be indicative of water quality well mentioned, through the installation of a 20 × 20 cm quadrat representing the minimum area. The observation stations were visited monthly, during a repetitive three-year cycle, during the spring, summer and autumn seasons, periods of maximum growth and development of the algal flora and the results suggest the following facts. The invasive alga Caulerpa cylindracea Sonder, 1845 tends to colonise disturbed ecosystems reflecting a reduction in native algal diversity; in fact, we note a drastic impoverishment of the invaded algal community, represented by a limited number of Macrophyte algae accompanying the invasive taxon in phytosociological surveys and a Shannon-Weaver Diversity Index (H’) and Equitability reduced by 4.49 and 0.77 n the heavily affected station. The number of macroalgal species accompanying the invasive species has dropped by 52% in Salamandre. In addition, the multidimensional analysis, represented by the Hierarchical Ascendant Clustering applied to this case, confirms our results.


Introduction
Studies on ENI date back to the 1970s (Ojaveer et al. 2018). Since then, global research in this area has developed very rapidly (Giakoumi et al. 2019).
The Mediterranean Sea is today considered to be one of the most affected of the regional seas, due to the various anthropogenic pressures on the different ecosystems of the Mediterranean asin (Mannino et al. 2017). This favors it to be an important place for nonnative species, which can become invasive leading to a loss of biodiversity and changes in ecosystem services (Brunel et al. 2013, Giakoumi 2014. The presence of mud-flats, estuaries and coral reefs along lagoons and rocky beaches provides ideal habitat for sustainable algae growth (Premarathna et al. 2019).
The concept of invasive species introduction is applied when the naturalisation of a species outside its usual geographic range is directly or indirectly linked to a human activity (Carlton 1985). The impact of introduced species has, thus, become the second most important cause of biodiversity loss in the world (Wilson 1997), after the outright disappearance of habitats. Bright (1998) defines the impact of introduced species on the planet's biodiversity as "evolution in reverse".
Amongst these bioinvasions, we are interested in Caulerpa cylindracea (Sonder 1845), recently reported on the Algerian west coast and which presents a strongly invasive behaviour during its colonisation (Manconi et al. 2020). It is considered to be one of the most threatening invasions of the Mediterranean Sea (Piazzi et al. 2016) and its impacts on algal biodiversity and seagrass beds are well reported in literature (Gribben et al. 2018, Montefalcone et al. 2015.
Indeed, the algal flora of the Algerian west coast remains largely unexplored. Direct observation of the algal settlement constituting the phytobenthos and its evolution is essential for monitoring possible changes in algal biodiversity. Therefore, the purpose of this research is to: • establish a quantitative and qualitative inventory, located in time and space, of macrophyte species inhabiting two sites located on the coast of Mostaganem (west coast of Algeria Fig. 1), one affected by C. cylindracea, the other spare. • assess the impact of invasive species expansion on taxonomic and specific diversity, abundance and dominance of algal groups, presence and/or absence of species of ecological interest that may or may not be indicative of water quality.

Material and methods
The choice of harvesting stations was primarily based on the importance of the algal marine flora, the location of the stations studied shown in Fig. 1.
The area studied extends over 50 km of coastline, from Salamandre each to Sidi Lakhdar each (small port). Both stations were visited monthly during a repetitive cycle of three years, in spring, summer and autumn seasons, periods of maximum growth and development of the algal flora. In each station, we prospected a 50 to 100 m long and 10 to 15 m wide coastal line, with a depth of 5 to 10 m.

Salamandre Station:
The Salamandre site ( Fig. 1) is 5 km from the town of Mostaganem, housing a new fishing port, its length around 4000 m. Surveys conducted in sheltered, quiet areas within a rocky beach with the following coordinates: 35°55' N/ 0°03' E.

Small Port Station:
This zone is located 35 km east of Mostaganem ( Fig. 1), the sampling stations located in places with sandy and rocky bottoms on a beach 3500 m long sheltering a clear-seeded Posidonia oceanica meadow, associated with macrophyte algae with geographical coordinates, 36°12' N / 0°23' E.

Sampling, sorting and conservation of material.
Sampling was performed on both soft and hard variable substrates. Sampling random, raking the rocks at each survey, three surveys per observation station and per year, according to climatological conditions and without biotope delimitation so that the sample is as representative as possible of the zone (Fig. 2). The geographical location of the studied sites (Google Earth modified).

Minimum Area
The minimum area is the area where one has the best chance of finding all the species of the settlementstudied to within 10% according to Gounout (1969). For our case, the minimum area adopted is 20 × 20 cm a surface of 400 cm Fig. 2.

Analytical Parameters
For the quantitative and qualitative analysis of vegetation, we used the terrestrial phytosociological methods that several authors have adopted in the marine environment (Boudouresque 1971a, Boudouresque 1971b, Cormaci et al. 1997 Figure 2. Quadrat used for sampling (20 × 20 cm). 1993). The results are expressed by their average, with data analysis carried out by the Statistica software (version 7.0) for all the parameters retained.

Recovery
The recovery (Ri) is the approximate percentage of the substrate surface covered in projection by species i. Given the stratification of vegetation, the total coverage of a survey ∑Ri is generally greater than 100%. The importance of recovery is expressed in class according to the following scale: + = negligible recovery 1 = less than 5% of the surface is covered 2 = between 5 and 25% of the surface area 3 = between 25 and 50% of the surface area 4 = between 50% and 75% of the surface area 5 = more than 75% of the surface area

Overall average recovery
Each class of the recovery coefficient Ri is assigned a conventional monthly average value (class centre) called average recovery. Absence = 0; + = 01%; 1 = 2.5%; 2 = 15%; 3 = 37.5%; 4 = 62.5%; 5 = 87.5% The RMG (overall average recovery) of species i in a set of N recorded is the average of these successive average recoveries.

RMG = ∑Ri/N
The RMG of a subset E of n species (ecological group, systematic unit or phytogeographic elements) is the sum of the constituent species.

Species richness (Q).
The Q coefficient is the specific population size of any subset in a survey (ecological group, systematic unit or biogeographic features) and the Q (highlighted) of a group of species representing its average species size.

Quantitative dominance.
The dominance based on recovery (∑DRi) of a group of species in a survey (in a survey table), is the ratio, expressed as a percentage of the sum of their recovery (or their RMG) to the total recovery of species in the survey (or to the total average recovery of species in the survey table).

Qualitative dominance (DQ)
This is the ratio expressed as a percentage of Q, (where Q highlighted) to T (where T highlighted).

Index of Specific Diversity (Shannon-Wiener)
This ndex (H') measures, in a settlement, the amount of information resulting from species differentiation. The H' value reflects the degree of structural evolution, maturity and stability of the ecosystem under consideration. It was calculated from the dominance of each species (Ri/Rt) according to Shannon's formula (in Boudouresque and Cinelli 1971

Taxonomic diversity
Three taxonomic indices have also been adopted for the study of taxonomic diversity (∆, ∆* and ∆+). These were calculated from a taxonomic tree constructed on the basis of phylogenetic classification , considering eight taxonomic levels: species, genus, family, order, class, phylum, under-reign and reign. From this tree, a taxonomic distance ωij is quantified between each pair of species, assuming that this distance is equal to 100 for two species related to the highest taxonomic level considered in the study (Clarke and Warwick 1999). This calculated distance (ωij) will then be integrated into the calculation of the various taxonomic indices. The calculations were performed using Suppl. materials 3, 1, 2 software R version 3.0.3 (R Core Team 2014). The phylogenetic classification is based on the Worms Editorial Board (Guiry and Guiry 2020).
With xi (i = 1, ..., S): the abundance of the i species, N (= Σ xi): the total number of individuals in the sample and ωij: the distance to be covered between species i and the first common node with species j in the hierarchical classification: -∆"Taxonomic diversity" (Warwick and Clarke 1995): 2 max 2 max th ∆ represents the average taxonomic distance between each pair of randomly selected individuals in the sample.
-∆*"Taxonomic Distinctness" (Warwick and Clarke 1995): ∆* is the average taxonomic distance between two randomly selected individuals belonging only to different species.
-Funnel Test: In order to detect assemblages whose taxonomic diversity would be influenced by disturbances,  proposed the "Funnel" test. This test consists comparing the taxonomic variety of a site to the area to which it belongs, from 1000 random draws of size m species made on the global list of S species listed in the area under consideration.

Hierarchical Ascendant Clustering (HAC)
The hierarchical classification has been used in a wide variety of disciplines. It has been described by several authors including Daget 1979, Dagnelie 1986, Legendre and Legendre 1984, Legendre and Legendre 1998, Herrera and Legac 2002and Bouroche and Saporta 2005. Our study was confirmed by the use of this multidimensional analysis which consists in partitioning the objects (or descriptors) of the study into groups and subgroups, passing through the condensation of the information provided through the data matrix into a similarity or distance matrix. The latter will then be classified according to different classification algorithms (Ward, single, complete, average), which will each provide a corresponding dendrogram. The calculation procedure for this analysis was performed using Suppl. materials 1, 2 under software R version 3.0.3 (R Core Team 2014).

Results
Results are reported in Table 1 and Table 3 for algal stand composition at the Salamandre and Small Port Stations respectively and in Tables 2, 4 for analytical parameters (RMG, DRi, Q̄ DQ) at the same stations. Table 5 represents the set of taxonomic indices at the two study stations and Fig. 3 illustrates the Funnel Test. The Hierarchical Ascendant Classification of the study stations based on the algae stands surveyed is illustrated in Fig. 4.  Table 2.
Analytical parameters at the Salamandre Station.   The results shown in Table 1 and Table 3, with illustrations and descriptions of sample specimens for the two observation stations are contained in a doctoral thesis ( Bachir Bouiadjra 2012).

Salamandre Station
The reading of the sheet on the algal settlement of the Salamandre site (Table 1) and the result of the structural parameters shown in Table 2, indicate a quantitative and qualitative dominance, respectively of 52% and 46% with an overall average recovery exceeding 50% of the species belonging to Chlorophyceae. Within this systematic unit, we note an overall average recovery rate of five invasive taxa, C. cylindracea, C. fragile, A. armata, F. rufolanosa L. lallemandii, which is of the order of 24.5%, 10.5% of which is reserved for the C. cylindracea, which is progressing in some surveys by joining its small colonies (Figs 5,6) to which are added the eutrophication indicator species of the environment and nitrophilic strongly accumulating nitrogen are represented by the order of Ulvales in the number of 8 species (Table 1) with a global average cover rate of 18% on a total global Table 5.
Taxonomic diversity indices. cover of 50% of Chlorophyceae in the Salamandre Station. This high presence rate is an indicator of a polluted environment which is confirmed by domestic and urban discharges (Khiari et al. 2017, Damak et al. 2019) duly observed at Salamandre Station, in addition the presence rate of the invasive species C. cylindracea amount to 10.83% (Table 1) which confirms an increasingly advanced colonisation.
The algal population of Phaeophyceae occupies a negligible average cover rate of 2.85% at the Salamandre Station and reveals the total absence of macrophyte species sensitive to pollution by detergents and water turbidity such as cystoseires (Cystoseira stricta, C. algeriensis, C. crinita ... etc) (Gamulin-Brida et al. 1967, Bellan-Santini 1966, Verlaque and Fritayre 1994a which confirms that this site is strongly impacted by C. cylindracea and disturbed, given the strong presence of the invasive species On the other hand the site of the small harbor displays a rate of recovery of Phaeophyceae macrophytes of 42.61% (Table 4) with a presence of nine species of Cystoseires suggesting a good water quality and an average algal specific richness of 76 species (Table 3) per survey contrary to the  previous site, which displays an average number of species of only 43 per survey (Table 1).
As for Rhodophyceae the overall cover rate is respectively 45.11% and 56.10% at the Salamandre and small port Station with a qualitative dominance of 44% and 43% (Tables 2, 4) which suggests protection and valorisation of some taxa rich in mineral elements and chemical components (Sambusiti et al. 2015, Cardoso et al. 2014, Shobana et al. 2017 see Ellisolandia elongata, Hypnea musciformis, Gelidium crinale, Gelidium spinosum, Gelidium corneum, Jania longifurca, Jania rubens and Gracilaria bursa-pasoris. This is easily verified in the number of species per phytosociological survey, which is much lower than those reported in the Small Port site and the reduction in the Diversity and Evenness Indices, respectively by 4.49 and 0.77 at the Salamandre Station, in relation to the results obtained at the Small Port. In this station invaded by C. cylindracea, a native species, Caulerpa prolifera (Forsskal) Lamouroux are recorded in surveys, with a low recovery rate of 2.5% compared to that of the invasive taxon (10.5%) suggesting a tendency for C. prolifera to be replaced by the invasive species (Piazzi and Ceccherelli 2006). Over time, according to , Jaubert et al. 2003, the homogenisation of the underwater landscape observed in some Mediterranean regions heavily affected by the invasive species has resulted. It is important to remember that for the four other invasive species of Rhodhophycea inventoried and subservient to the photophilic infralittoral of hard substrates, seem to find favorable conditions for their development, without any threat to the ecosystem. Nevertheless, the Bonnemaisonncea of the genus Asparagpsis found in Salamandre, has major assets its avoidance by herbivorous fish because secreting secondary metabolites and its ability to multiply vegetatively and by apomeiotic spores (Feldmann 1954) hence its expansion to the site of Salamandre with a recovery rate of 1%, thus forming a settlement in Asparagpsis.
In addition, the beach of Salamandre with coastal developments operating in recent years is a vulnerable site to closely monitor to avoid erosion of its algal richness due to an expansion of invasive taxon C. cylindracea.

Small Port Station
It is a site rich in species, 109 taxa listed (Table 3), 80% of the total number recorded on the coast of Mostaganem ( Bachir Bouiadjra 2012). The recovery rate of the entire algal flora exceeds 100%, with a Diversity Index (H') of 5.56 which is higher than that of the previous station (Salamandre), as well as that of Bordj El kifan (1.62) east of Algiers reported by Serridi (2003). The Evenness Index is about 0.82, which suggests, according to Daget (1979), a balanced algal stand.
The overall average overlap among Rhodophyceae is clearly dominant at 56%, the same observation is made for their quantitative and qualitative dominance estimated at 43%, followed by the Phaeophyceae group with an RMG of 42% and dominance (Dri) and (DQ) of 32% and finally the Chlorophyceae group with an RMG of 31% (Table 4). This represents a well-structured settlement with species constituting the crusty, grassy, erect and epiphytic strata according to the algal settlement sheet (Table 3). Note the absence of the invasive taxon C. cylindracea however, the fishing port adjacent to the observation sites, will constitute a potential vector for caulerpe transmission in the area. Hence the need for increased vigilance and follow-up with awareness campaigns among fishing professionals to avoid contamination of the site.

Taxonomic Diversity Indices
Analysis of the different taxonomic Diversity Indices (Table 5) indicates a higher taxonomic diversity at the Small Port than at the Salamandre Station. This means that the species recorded at the Small Port are more taxonomically distant than those found in Salamandre.
In other words, the average taxonomic distance between each pair of randomly selected individuals is greater at the Small Port than at Salamandre.
The Funnel Test (Fig. 3) based on the values of ∆+ obtained for each Station, in comparison with the mean expected value in the study area, shows that the Salamandre Station is outside the 95% confidence interval of ∆+m, indicating a low taxonomic diversity compared to that expected in the area. This suggests that this site is subject to anthropogenic disturbances, influencing the taxonomic composition of the settlement by generating a poor assemblage of species, in addition to the presence of invasive species in this plant, associated with untreated wastewater discharges, confirm this reduction in algal diversity (Verlaque and Fritayre 1994b, Piazzi and Cinelli 2003, Piazzi and Ceccherelli 2006.

Hierarchical Ascendant Clustering
Our study was confirmed by a multidimensional analysis which is the Hierarchical Ascendant Clustering (HAC). The latter (Fig. 4) shows a clear separation between the two sites considered by indicating four distinct groups of phytosociological surveys. The first two, represented by the summer and spring surveys at the Small Port Station, are the most dissimilar of the other groups and represent the greatest distance. The two surveys, therefore, represent the most species-rich removal recorded, indicating 89 and 72 species recorded respectively during the spring and summer seasons at the Small Port. However, we note a decrease in the number of species in the surveys conducted during the winter season at the same site with 68 species recorded, which results in a smaller distance in the classification tree for this survey. The last group consists of the surveys carried out in the Salamandre Station and represents the smallest distance in the dendrogram due to the low number of species recorded due to the dominance of the settlementby species belonging to Chlorophytes, neutrophils associated with five other invasive species including C. cylindracea, C. fragile, A. armata, F. rufolanosa and L. lallemandii, which testifies to the disturbance and degradation of the environment in the Salamandre Station.

Conclusions
The invasive alga Caulerpa cylindracea Sonder, 1845 tends to colonise disturbed ecosystemseflecting a reduction in indigenous algal diversity , Piazzi and Cinelli 2003, Piazzi and Ceccherelli 2006. This is confirmed in the heavily colonised observation site of Salamandre Beach, through the statistical analysis tools used to see the parameters of abundance recovery, qualitative and quantitative dominance of algal groups, Equitability Index and taxonomic diversity and where there is a drastic depletion of the invaded algal community, represented by a limited number of Macrophyte algae accompanying the invasive taxon in phytosociological surveys and a low Shannon-Weaver Diversity Index (H) of 4.49. On the other hand, the Station of the Small Port not affected by C. cylindracea hosts a population exceeding in some surveys about twenty taxa or the Diversity Index is 5.56 suggesting a well-structured and stable algal population. The number of macroalgal species accompanying the invasive species has dropped by 52% in Salamandre. Verlaque and Fritayre (1994b) attribute this decrease to a rate ranging from 25 to 55%. This is consistent with our results. Furthermore, the analysis of the floral procession mentioned in the surveys of the Salamandre Station indicates a decrease in the cover of erect shrub algae (-70 to -80%) and epiphytic and filamentous algae in the autumn season, while encrusting algae of the genus Corallina dominate and best resist the occupation of the ecological niche by the invasive species C. cylindracea. This study provides an inventory of the algal population associated with the recentlyintroduced C. cylindracea at the Mostaganem coast, which needs to be monitored at different levels of the affected coastline in order to effectively limit its expansion to the detriment of existing algal diversity.