Influence of Zostera marina L. and Z. noltei Hornem. (Zosterales: Liliopsida) on the composition, species richness and coenotic diversity of the phytobenthos of the Azov Sea

The influence of marine higher plants Zostera marina and Z. noltei on the composition, species richness and coenotic diversity of the phytobenthos of sandy and silty soils of the Taman B y in the Azov Sea at the depths from 0.5 to 3 m has been investigated. I t is shown that 1) macrophytobenthic communities with a high degree of Z. noltei and Z. marina dominance (DD), both at very small and relatively large sites, are characterized by lower species richness than the co mmunities with low DD; 2) that being said, for the majority of co-occu rring species it is possible to grow in coenoses with any DD; 3) accord ingly, about the same number of species can grow at the small community s ites with low DD and at the larger sites with high DD; 4) DD has little o no influence on the species diversification of coenoses. However, both the growth of Zostera marina and Z. noltei and the decrease in their participation can lead t o considerable drop in occurrence of some macroalgae sp cies or to disappearance of those species from a community.


Introduction
Dominant species are considered to be an important internal factor that regulates species richness in plant communities [1]. However, for the reason that the manner, in which various dominant species influence communities of various types, can differ substantially, the knowledge concerning this aspect of plant cover structure is still tentative and vague, especially when considered in the context of different spatial scales.
Thuswise, dominant species can affect other (co-occurring) plant species, limiting their space or mineral supply. Such manner of influence entails non-selective (random) displacement of other species, which means that any species has a certain probability of occurring at any site within a coenosis with any participation of a dominant species (a degree of species dominance, or DD). Accordingly, about the same number of species can grow at a small community site with low DD and at a larger site with high DD.
However, there are other ways, in which dominant species can exert an impact on cooccurring species -for instance, by means of ecotope transformation or allelopathy [2,3]. In this case, some species can prove to be more susceptible to such influence than the other ones, and, when dominant species strengthen their position, some co-occurring species may completely disappear from large sites of plant communities. Additionally, in the communities of extreme habitats, both terrestrial and aquatic (sea bottom), mutually favorable relationships between plant species are a common occurrence. For this reason, deterioration of the dominant species populations in such coenoses may lead to the decrease in the species richness, and not vice versa [3][4][5][6].
The Azov Sea is characterized by low and unstable salinity (5-15 ‰) and considerable turbidity of its water. Such conditions are detrimental for the formation of macroalgae communities, so their propagation here is very limited. The main dominant species in these coenoses are flowering plants Zostera noltei and Z. marina [7]. As these species comprise the major share of the biomass of macrophytobenthos, it may be assumed that they have a significant influence on its species composition and coenotic diversity. However, relevant information in the literature is virtually absent. This work is aimed at filling this gap in knowledge.

Materials and Methods
This study is based on 165 samples of the macrophytobenthos collected in the Taman Bay of the Azov Sea in summer, 2020; the sampling was conducted in 6 homogenous areas of coenoses with dominance or participation of Z. noltei and Z. marina, located at the depths ranging from 0.5 to 3 m near settlements Sennoy, Taman and Garkusha (coordinates: 44°65.150'-44°80.786'N, 37°37.065'-37°77.264'E). Within each selected community site, 25-30 sampling plots of 0.5× 0.5 m size were set; some of them were set in a regular pattern shaped as one or two transects comprising 10 plots each; the other ones were set in series, 5-15 plots per a site. In the second case, the investigated communities were chosen to have options with high and low total projected cover of Zostera, which was assessed visually. All macrophytes from each frame were collected in a separate gauze bag. Each sample was sorted out by species, dried with absorbent filter paper and weighted, separately for each species [8]. Algae were identified according to identification guides [9,10] with regard to the most recent changes in their nomenclature [11].
In order to assess and characterize the influence of Z. noltei and Z. marina on the species richness of macroalgae communities and on the species' frequency of occurrence, the samples were ranked based on DD (from the lowest to the highest) and divided into four equal groups. For each group, the average number of co-occurring species in the sample (provisional local species richness), the total number of such species in all samples of the group (provisional regional species richness), and their frequency of occurrence have been estimated, and the values of these parameters were then compared.
To assess the influence of Z. noltei and Z. marina on the diversification of species composition in the communities (beta diversity), we estimated species similarity between all six sites; it was done separately for the groups of samples with the lowest and the highest DD (10 samples per each group). For this purpose, the Sørensen coefficient was used (Ks = 2С/(А+В), where A and B are the number of species in 10 samples with the lowest (or the highest) DD, collected at the sites under comparison; and C is the total number of species in the groups of samples collected at the sites under comparison). It is assumed that the lower is the similarity coefficient, the higher is the diversification of a community in terms of its species composition.

Results
The samples with differing Zostera participation are characterized, on average, by significantly differing species richness in 0.25 m2 (ANOVA, F2.66 = 11.71, P= 0.05): the higher DD is, generally the lower are the values of SS, average for the group. It should also be noted that three groups of the samples with DD > 0.88 are characterized by a similar total number of species (Ns = 22, 17, and 19), which is significantly lower than the total number in the group of samples with DD < 0.88 (30 species). However, when the total number of species in all the samples with DD both more and less than 0.88 (41 and 122 samples with the total areas 10.3 and 30.5 m 2 , respectively) is considered, it appears to be similar: 30 and 27 species.
The comparison of the frequencies of occurrence of macroalgae species in the groups of samples with different DD has shown that 23 species occur in the samples both with relatively low and very high DD (among them, 6 species have a considerable higher frequency of occurrence in the coenoses with low DD; only Ruppia maritima shows a much higher frequency of occurrence in the coenoses with high DD). Seven species occur only in the samples with a relatively low DD, and four species are found only in the samples with a very high one. Some species from the second group are rare to occur in Zostera communities in general. For instance, Ulva rigida actively propagates in eutrophic waters, while Zostera is averse to such environment. Stuckenia pectinata prefers fresh waters, and it is uncommon for it to enter the waters with the salinity exceeding 5-7 ‰, which is the opposite for Zostera. Therefore, it would be unwarranted to assume that the absence of these species in Zostera thickets is the result of their competition for space and/or resources. It is more plausible that their distribution is influenced by other factors. As for the species in the first group, an increasing share of Zostera participation can threaten the well-being of Charophyceae Lamprothamnium papulosum and Chara baltica. They are negatively affected by the increase in density of Z. noltei и Z. marina thickets, and have been recorded only in isolated sites of the investigated area.
The average values of the Sørensen coefficient of species similarity between the sites of coenoses with low Z. noltei and Z. marina participation (DD = 0.69±0.04) were 0.49±0.04, n = 15, and for the ones with high DD (DD = 0.99±0.00) they were 0.56±0.04, n = 15. That is, the sites of coenoses with high DD are slightly more homogenous in terms of species composition than the ones with low DD, however this difference is not statistically significant (ANOVA, F4.20 = 1.27, P= 0.05).
As it follows, the results of this investigations show that 1) macrophytobenthic communities with a relatively high Z. noltei and Z. marina participation, both at very small and relatively large sites (in the groups of biomass samples), are characterized by lower species richness than the communities with relatively low DD; 2) for the majority of cooccurring species it is possible to grow in coenoses with any degree of Zostera participation, and the frequency of occurrence of the vast majority of them has next to no association with Z. noltei and Z. marina; 3) about the same number of species can grow at the small community sites with low DD and at the larger sites with high DD; 4) the sites of coenoses with high and low DD are characterized by a similar degree of diversification of their species composition. However, both the growth of Z. marina and Z. noltei and the decrease in their participation in coenoses can lead to the decrease in occurrence of some