Diatom Species that Characterize Saline Ponds (Southern Spain) with the Description of a New Navicula Species

The southern Iberian Peninsula has a high number of saline ponds where electric conductivity (EC) is an important factor that directly affects the distribution and abundance of aquatic organisms. Environmental factors (such as pH, EC, and temperature) were measured, and diatom assemblages were sampled in 15 shallow saline ponds in southern Spain over a range of EC (1.4 mS to 51.6 mS cm−1). Three groups of ponds were defined based on EC (oligosaline 1.4 to 5.3 mScm−1, mesosaline 10.9 to 17.3 mScm−1, and eusaline 32.3 to 51.6 mScm−1), and sediment diatom assemblages were studied. PERMANOVA analysis revealed significant differences in diatom community composition between the three groups of ponds. Non-metric multi-dimensional scaling analysis (n-MDS) showed distinct clusters of diatom assemblages in oligosaline and mesosaline ponds. The dominant diatom species in the eusaline ponds were Tryblionella pararostrata (Lange-Bertalot) Clavero & Hernández-Mariné, Halamphora cf. pertusa J.G. Stepanek & Kociolek, Halamphora sp.1, and Cocconeis euglypta Ehrenberg; the mesosaline ponds were dominated by Navicula veneta Kützing, Nitzschia elegantula Grunow in Van Heurck, and Planothidium delicatulum (Kützing) Round & Bukhtiyarova; and the oligosaline ponds were dominated by Navicula veneta, Pseudostaurosira brevistriata (Grunow) D.M. Williams & Round, and Nitzschia inconspicua Grunow. A new diatom species was described from three eusaline ponds (32-51.6 mS cm−1). A detailed description of N. maiorpargemina sp. nov. is presented in this study based on light and scanning electron microscopy after comparison with morphologically and ecologically related taxa.


Introduction
In recent decades, different governments and international organizations have pursued efforts to conserve and protect wetlands through the Ramsar Convention as the first global treaty to protect nature González-Capitel 2002, Elvan andBirben 2021). According to the Ramsar Convention (2016), wetlands include estuaries, marshes, lagoons, lakes, and ponds. Aquatic ecosystems can be classified according to the number of ions found in them, and the term "saline" is used for continental waters rather than "haline" as indicated in different studies (Gasse 1987;Reed et al. 2012). Particularly, saline ponds are sensitive aquatic ecosystems, as they reflect the consequences of human and natural disturbances (Waiser and Robarts 2009;Potapova 2011). Endorheic systems, such as some saline ponds, are mostly small (< 50 ha) and shallow (< 1 m) (Stenger-Kovács et al. 2014) and, being relatively closed systems, they constitute reliable sentinels of climate change (Gottschalk and Kahlert 2012).
This article belongs to the Topical Collection: Wetland Algae and Cyanobacteria.
Different studies in saline aquatic ecosystems have indicated that the increase in ionic concentration in the water has a progressively negative effect on the number of species (Hammer 1990;Amores et al. 2013). These extreme environments are highly species selective, causing species lacking osmoregulatory mechanisms to disappear (De Deckker 1988). One way that an organism can cope with salinity is to compensate for the high ionic concentration by accumulating organic solutes in their body (Waiser and Robarts 2009). Halotolerant/halophilic organisms can inhabit these environments and therefore a high number of rare or endemic species are usually present (Millán et al. 2011).
Despite this, different groups of organisms such as bacteria, fungi, protozoa, zooplankton (copepods, rotifers or cladocerans), microalgae and macrophytes (Ruppia, Carex, Salicornia) are adapted to high levels of NaCl (Waiser and Robarts 2009). Species belonging to the unicellular flagellate green algae Dunaliella have shown high NaCl concentrations in the cell cytoplasm that can accumulate glycerol as an adaptation to hyperosmotic environments (Schubert et al. 2010;Lowenstein et al. 2011;Oren 2014). Other genera such as Chlorella, Scenedesmus, Desmodesmus or Monoraphidium have also shown tolerance to high salt concentrations (Figler et al. 2019).
Microphytobenthos, which are composed mainly of cyanobacteria, green algae, and diatoms, colonize saline ponds (Stevenson 1996;Waiser and Robarts 2009) and are the main carbon fixers in shallow ponds (Montoya 2009). Photosynthetic activity of the epipelic microalgae make up about 30% of total primary production, especially in the deeper infra-littoral or mid-depth zones (Cantonati and Lowe 2014). Diatoms can be tolerant to high salt concentrations by the active transport of ions out of the cell or into the vacuole (Shi et al. 2002) and by the regulation of cellular osmolytes (Clavero et al. 2000), although these mechanisms are energetically costly. An efficient and less costly mechanism is through the adsorption of salts to the extracellular polymeric substance (EPS) such as uronic acids or sulphates, which provide a competitive advantage to diatoms in salt-stressed environments (Allan et al. 1972;Abdullahi et al. 2006;Steele et al. 2014).
Ecological studies in peryphytic diatom assemblages in ponds/lakes are scarce in Europe, though some have been undertaken in mountain ponds of France (Beauger et al. 2020), coastal ponds in Italy (Della Bella and Mancini 2009) or saline lakes in Hungary Ács 1996-1997;Stenger-Kovács et al. 2014) and Romania (Nagy et al. 2008). Shallow and hypersaline endorheic ponds in Western Europe are mostly restricted to Spain (Reed 1998a, b;Woodward 2009;Casamayor et al. 2013). The saline ponds of Spain show wide ionic variety (Montes and Martino 1987), due to sediments and seasonal fluctuations along the year (typical of the Mediterranean climate) (Gasith and Resh 1999;Guerrero and Wit 1992;Martín et al. 2010;Lucena-Moya et al. 2012). Diatom studies in saline habitats such as rivers, lakes, estuaries, lagoons, and marshes show that most diatom genera tolerate high levels of salinity and conductivity (Potapova 2011).
Particularly, in south-eastern Spain, many studies have been made in saline ponds (Aboal 1989), in wetlands such as in Pego-Oliva marsh with different ranges of EC (Cantoral-Uriza and Aboal 2008), and in coastal and hypersaline lagoons (Belando et al. 2012;Antón-Garrido et al. 2013). From a palaeoecological approach, Reed (1998a, b) studied more than 50 saline ponds in southern Spain, including some ponds examined in the present study (e.g. Zarracatín, Gosque, Zóñar, Ballestera, Tíscar or Ratosa), where diatom communities were related to chemical variables such as conductivity. Potapova and Charles (2003) reported that EC is a major contributor to the distribution of diatoms, in agreement with Negro and Hoyos (2005) who reported EC and alkalinity as critical variables in the distribution of planktonic diatom communities in reservoirs of Spain. Some other diatom studies published inventories of the benthic diatom communities of the ponds at different sites of Spain (e.g. Armengol et al. 1975;Ubierna León and Sánchez-Castillo 1992;Blanco et al. 2004), as well as certain mountain ponds (Linares-Cuesta 2003; Rivera-Rondón and Catalán 2017), and new species were described for southern Spain (Sánchez-Castillo 1993;Blanco et al. 2013Blanco et al. , 2019.
The principal aim of the present study was to describe the diatom communities that characterize 15 saline ponds in the South of Spain, including the description of a new species Navicula maiorpargemina sp. nov. that appeared in three of the ponds studied.

Sampling Design
In this study, 15 shallow ponds were selected over a wide EC range (1.4 to 51.6 mScm −1 ) (Table 1) and sampled between April 2004 to June of 2006. The distribution map showing the sampling sites ( Fig. 1) was drawn using QGIS Dufour 2.0. Physical-chemical parameters such as pH, electric conductivity, and water temperature were measured in situ with a multiparameter probe (pH/Cond 340i Portable Multi-Parameter, WTW model). Other parameters, such as elevation, distance to the sea or maximum pond depth (Table 1) were determined from the official website of the environmental agency of Andalusia (Consejeria de Medio Ambiente 2020). According to prior studies in these wetlands (Cowardin et al. 1979;Lucena et al. 2012), and based on the conductivity range, the ponds were classified as oligosaline (< 8 mS cm −1 ), mesosaline (8 -30 mS cm −1 ), and eusaline (> 30 mS cm −1 ).
Several transects were made in the littoral zone in order to take an integrated sample of the epipelic community. A sediment sample from each pond was collected by suction through a one-meter-long glass tube using standard methods (Round 1953;Polge et al. 2010). Samples were transferred to plastic bottles and preserved with formaldehyde (4% final concentration) before laboratory treatment. Diatom samples from sediments were collected from these 15 shallow ponds.

Water and Diatom Sample Processing
In the laboratory, the sediment from each epipelic sample was left to stand in a Petri dish, then the supernatant was removed, and the sediment was covered with coverslips, which were removed at least 12 h later to collect diatoms attached to the coverslip by phototactic movement. The coverslips were washed to remove the diatoms attached to them. Samples were oxidized in glass tubes with hydrogen peroxide 30% w/w. Then, calcium carbonate inclusions were removed by adding a few drops of diluted hydrochloric acid 1 N to the sample. Samples were then washed with distilled water four times by centrifugation (3000 rpm, 5 min), the supernatant removed, leaving the clean pellet. Permanent slides were mounted with synthetic resin (Naphrax®) having a high refractive index (1.74).
Light micrographs of the diatoms were taken under a Zeiss Axioplan 2 imaging microscope with differential interference contrast (DIC) and a 1000x immersion objective (NA 1.40) equipped with an Olympus DP70 digital camera. Measurements of valve length, width and number of striae in 10 μm were taken of at least 30 specimens per species, under the light microscope. For scanning electron microscopy (SEM) analysis, metal stubs were prepared with organic-free samples air dried on a thin pellicle of graphite (EMITECH K 950X) and coated with gold palladium (Polaron equipment limited SEM coating unit E5000). All images were digitally edited, and plates were made using CorelDRAW.

Type Material Study
Navicula pargemina Underwood and Yallop was described from Severn Estuary in the UK and the holotype deposited in the Natural History Museum of London (BM-82,311) ( Fig. 2). Considering that the newly described Navicula species shows similarities with N. pargemina, we compared it with the original data of Underwood and Yallop (1994), Witkowski et al. (2000a, b) and by observation in this study under the LM of type material (Fig. 2). Morphological and morphometric characterization was based on measurements of 30 valves from this type material, which was used to compare with the new species described in the present study.

Data Analysis
The distribution patterns of the diatom communities were explored in the 15 ponds using a non-metric multi-dimensional scaling analysis (n-MDS) based on Bray-Curtis similarity matrix and cluster analysis with all samples. The relative-abundance values of the different species were used for analyses. Statistical differences between the three groups of ponds based on EC (as achieved by the n-MDS) were tested by a permutational multivariate analysis of variance (PER-MANOVA test). Moreover, a SIMPER analysis and cluster (double-square root transformed data on a Bray-Curtis similarity matrix) was used to find the most representative taxa for each statistically different salinity. We used Primer 6 and PERMANOVA + (Clarke and Gorley 2006).

Physical-chemical Variables and Salinity Gradient
Three different groups of ponds were found in this study in terms of EC. The physical-chemical variables of the 15 ponds measured in situ are shown in Table 1. The ponds with the lowest EC (oligosaline; 1.4 to 5.3 mScm −1 ) were Taraje, Zoñar, Rincón, Calderón, and Pilón. These ponds, at 45 to95 km from the sea and at elevations of 80-345 m a.s.l, had a pH range of 7.8-8.6 and water temperature between 22 and 31.7ºC.

Diatom Community Composition
A total of 150 taxa from 55 diatom genera were identified in the 15 samples from the ponds studied. The n-MDS ordination plot of diatom samples showed a stress of 0.14 ( Fig. 3a) and the cluster analysis ( Fig. 3b) showed the distance between diatom samples. A clear separation, based on diatom assemblages, between eusaline and oligosaline ponds is visible along axis 1 of the ordination diagram with the mesosaline ponds mediating the other two groups of ponds (Fig. 3a).
The PERMANOVA global test showed that the diatom communities in each type of pond had statistically significant differences according to EC (Pseudo-F = 1.513. p = 0.012.). Pairwise tests revealed significant statistical differences between eusaline and oligosaline as well as between eusaline and mesosaline ponds (Table 2). Nevertheless, the communities did not significantly differ between mesosaline and oligosaline ponds (t = 0.952, p = 0.652).
The SIMPER analysis revealed a dissimilarity of 98.37% within diatom assemblages from eusaline and oligosaline ponds and 94.48% within diatoms from eusaline and mesosaline ponds (Table 3). Meanwhile the average dissimilarity between mesosaline and oligosaline was lower in the SIMPER analysis (87.78%).
The main species  show the differences that exist between ponds of different conductivities. Eusaline ponds were differentiated from the rest of the ponds by the following species: Halamphora cf. pertusa J.G.
Subclass Bacillariophycidae Round (Round et al. 1990 Etymology: Refers to the larger size when compared with the most similar species Navicula pargemina. Fig. 3 a Non-metric multidimensional scaling analysis (n-MDS) ordination plot of diatoms from ponds in southern Spain. Blue triangle: oligosaline ponds; green triangle: eusaline ponds; blue square: mesosaline ponds. b Cluster Analysis of diatoms from ponds in southern Spain. Blue triangle: oligosaline ponds; green triangle: eusaline ponds; blue square: mesosaline ponds  (Fig. 52).
Externally the raphe at the poles hooked (Fig. 53).

Comparison of the new species with type material of Navicula pargemina and other similar species
Morphological features of Navicula maiorpargemina sp. nov. and comparison with five species of Navicula which appear in high-conductivity waters are presented in Table 4. In this study, N. maiorpargemina appears in water with high EC, like some other small Navicula. Nevertheless, these have clear morphological and morphometrical differences which easily distinguish them in the same samples.
In N. pargemina (Underwood and Yallop, 1994), Figs. 11-12, the central area is wider than in N. maiorpargemina sp. nov. (Figs. 49-51). Hymens were not found in N. maiorpargemina sp. nov., as described in N. pargemina, and the number of lineolae was 60-70 for the new species instead of 100 counted in N. pargemina.

Discussion
Our work addressed the variation of diatom communities from pond sediments under different electric conductivity. The study sites constitute part of a more extensive monitoring network with more than 50 ponds from Andalusia, since they represent more than 25% of ponds in this area (provided by the Regional Government of Andalusia). The analysis of the diatom assemblages in pond sediments indicated differences between lower EC groups (oligosaline and mesosaline) and higher (eusaline). This fact has also been observed in Oliva-Pego wetlands where diatom species characteristic of brackish and freshwaters were defined (Cantora-Uriza and Aboal 2008).
The cluster analysis characterizes more clearly the similarity among the ponds studied, confirming that electric conductivity plays a critical role in the distribution of peryphitic diatoms, as observed in other studies (Veres et al. 1995;Potapova and Charles 2003;Tibby et al. 2007;Stenger-Kovács et al. 2014), Balance between evaporation and precipitation, especially during periods of severe drought (e.g. summer months), influences the variability in EC values (Veres et al. 1995), especially in the shallowest and most saline ponds such as Zarracatín (Table 1). This trend is consistent with the changes noted in other ponds, such as Gosque or Ratosa, included in the eusaline group. Nevertheless, the maximum in this study did not reach the previously reported highest values with more than 100 mS cm −1 in some years (Montes and González-Capitel 2002), this level pertaining to the hypersaline category. Mesosaline ponds such as Honda proved more variable, having periods with subsaline to hypersaline conditions. However, oligosaline ponds (e.g. Taraje, Pilón, Rincón, Zóñar) clearly had hydrological characteristics corresponding to those that Montes and González-Capitel (2002), categorized as subsaline ecosystems.
Although the genera of the most representative taxa found in the mesosaline to eusaline ponds from this study were represented mainly by Amphora sensu lato (which includes Halamphora), Cocconeis, Navicula, Nitzschia, Planothidum, and Tryblionella, the main genera were Navicula and Nitzschia, as reported by Stenger-Kovács et al. (2014) as more dominant in saline waters.
Tryblionella pararostrata (Lange-Bertalot) Clavero & Hernández-Mariné (in Clavero 2009), which is usually found in marine or hypersaline environments (Schoeman 1972;Siqueiros-Beltrones et al. 1991;Clavero 2009)  study and has also been reported in estuarine environments in Portugal (Ribeiro 2010) and is therefore considered a characteristic species of high-conductivity waters. Halamphora sp.1 is a small diatom that, together with Cocconeis euglypta Ehrenberg, filamentous algae, and carophytes such as Lamprothamnium papulosum (Wallroth) J. Groves was very abundant in the group of ponds with the highest conductivity (Gosque and Ratosa) and has frequently been reported in these ecosystems in Spain (Cantoral-Uriza and Aboal 2008).
Diatom species that characterized mesosaline ponds were Nitzschia elegantula Grunow in Van Heurck and Planothidium delicatulum (Kützing) Round & Bukhtiyarova 1996, found in Salada pond with more than 10 mS cm −1 . The species Nitzschia elegantula has been reported in other studies as halophilous and dominant in media with moderate electrolyte content (Della Bella et al. 2007;Solak et al. 2019). In Spain, N. elegantula has been reported as characteristic of brackish waters (Antón-Garrido et al. 2013).
The oligosaline group was characterized by the presence of Pseudostaurosira brevistriata (Grunow) D.M. Williams & Round 1988, which was the most abundant diatom species and apparently more sensitive to higher EC values (Tibby et al. 2007) although this species tolerates a wide range of conductivity (Antón-Garrido et al. 2013). Nitzschia inconspicua    Grunow and Navicula veneta Kützing were also dominant in these ponds, especially the latter, which is considered by different authors as a "salt generalist" .
Other taxa found in this study such as Navicula cf. jakhalsensis Van Landingham, in Clavero et al. (2000), and Caloneis permagna (Bailey) Cleve, in Zarei-Darki (2011) and Resende et al. (2005) (Appendix 16), were reported as mesohalobian in estuarine habitats, respectively, and were also found in the present study in the eusaline ponds but with lower abundance compared to other taxa mentioned above. In Cantoral-Uriza and Aboal (2008) and Antón-Garrido et al. (2013), the species Navicymbula pusilla (Grunow) Krammer is reported as abundant in high-conductivity waters. In the present study, it appears in Consuegra pond (Appendix 16), which is in the mesosaline group, but it has not been identified in the eusaline group, which might represent the limit for conductivity tolerance of this species.
According to the literature, the most similar species to N. maiorpargemina is Navicula pargemina (e.g., Underwood and Yallop 1994;Clavero et al. 2000;Ribeiro 2010). A taxon identified as Navicula cf. pargemina has been recorded in the Loire estuary by Ribeiro et al. (2019). The new taxon described here presents clear differences with respect to Navicula pargemina, such as larger size, a lower number of striae, and a more rhombic-lanceolate shape. The comparison between Navicula maiorpargemina sp. nov. and other similar taxa reveals a unique combination of characters, such as valve shape and an asymmetrical central axial area, whereas the valve pairs after oxidation are the most characteristic trait of this species, found in no other except in N. pargemina. Therefore, it is probable that N. maiorpargemina sp. nov. is more common in European estuaries (Ribeiro 2010;Ribeiro et al. 2019) and saline ponds than so far reported.
The environments with high conductivity are highly sensitive to climate change (Hammer 1990;Jeppesen et al. 2015), and diatom communities are affected by this global change (Reed 1998a, b). Detailed analyses of the long-term dynamics in ponds as well as key factors and interactions affecting water quality (Ventelä et al. 2016) are needed for better management, and conservation measures are enacted in these sensitive environments. More studies concerning the autoecology of diatom species in these wetland environments are necessary to establish the optimal and tolerance ranges of the different species. Table 5   Table 5 Relative abundance (> 5%) of taxa in ponds