In the northern and northwestern part of the Northern Hemisphere, a significant fraction of lakes is represented by small and mostly shallow lakes located on the acidic substrate and fed mainly by precipitation. This makes them soft-water lakes, poor in dissolved salts of mainly calcium (< 7.0 mg Ca l−1), usually acidic (pH 4.5–6.7) or only slightly alkaline, oligo- or mesotrophic and highly vulnerable to human activity (Murphy 2002). In some of these lakes, we can observe isoetids (a group of aquatic macrophytes), such as
In Europe, lakes with isoetids are located on the Scandinavian Peninsula, the British Isles, in France, the Netherlands, Germany and Poland (Hultén & Fries 1986). The EUNIS habitat classification system lists them under code 3110. In Poland, there are about 200 lakes with isoetids (Szmeja 1996) located in Pomerania (NW Poland) in the area of end and ground moraines and outwash. The phytolittoral substrate of such lakes is acidic, mineral and organic, highly hydrated, fairly well oxygenated and lighted, usually free from wind wave pressure or other hydrodynamic forces, located at a depth of 0.3–6.5 m (2.0–5.0). In Poland, most lakes with isoetids are characterized by their location in forest and are protected by law, while others are located in close proximity to cultivated land or buildings and are exposed to intense human activity (Szmeja 2006).
Research on lobelia lakes has a long history and relates mainly to ecophysiology (Sand-Jensen 1978; Sand-Jensen & Borum 1984; Madsen 1985; Boston 1986; Boston et al. 1987) as well as habitat requirements of isoetids (Seddon 1965; 1972; Mäemets 1974; Rørslett 1991; Vöge 1992; Szmeja et al. 1997). There are few studies exploring the structure of submerged vegetation and its transformation caused by environmental factors and anthropogenic pressure.
Specific environmental conditions in lakes with isoetids are subject to alkalinization (Arts 2002), affecting the appropriate carbon content for photosynthesis and disrupting the metabolism (Sand-Jensen 1978; Pulido et al. 2011; Chmara 2014). Our hypothesis is that alkalinization of lakes is responsible for plant species replacement. The exploration of such lakes to date (Chmara et al. 2015) indicates that acidophytic mosses and isoetids may be replaced by some vascular plants and charophytes. Identification of a plant replacement trend that would closely correlate with environmental characteristics of the aquatic environment affected by human impact may contribute to more effective biodiversity conservation within this group of lakes throughout their geographic range.
The study was conducted between 1998 and 2018 in 70 soft-water lakes of northwestern Poland (Fig. 1), where 12 702 plant samples and 38 106 water samples were collected (the latter were collected in places with plants). In each of these lakes, one transect perpendicular to the shoreline was delineated and divided into strips parallel to the shore at 1 m depth intervals. They are referred to as depth zones throughout this study (Fig. 2). In each of the depth zones, a diver randomly delineated 10 to 20 square fields of 0.1 m2, registered all plants present in these fields, determined their abundance (expressed as % of a sample). Data from such fields are called plant samples in this study.
From each of the bottom zones or depth zones with 1 m intervals, the diver collected three water samples (a total of 1.5 dm3) 5 cm above the plants or sediment. The water samples were analyzed chemically and physically according to the methods proposed by Hermanowicz et al. (1999) and Eaton et al. (2005). The concentration of calcium, total nitrogen, total phosphorus and humic acids, as well as the hardness of water and its color, pH, redox and conductivity were determined. During the fieldwork, photosynthetically active radiation (PAR), water clarity, water temperature and oxygenation were measured.
The environmental factors were determined in the depth zones as per the following parameters:
depth (m) – using NEXUS DEPTH or Eagle TriFinder depth finders;
rection – pH-meter 320/SET1 with a SENTIX 97T measuring electrode;
water clarity (m) – Secchi disk with no glass pane on water surface;
calcium concentration (mg Ca dm−3) – complexometric titration, titrating 50 dm3 of water against EDTA disodium, with calconcarboxylic acid as an indicator;
water oxygenation (%) and temperature (°C) – WTW OXI 197 oxygen meter with an EOT 196 electrode;
general water hardness (mg CaO dm−3) – complexometric titration with Eriochrome Black T as an indicator;
humic acid concentration (mg KH dm−3) – spectrophotometrically, using a UV-VIS Aquamate spectrophotometer at 330 nm wavelength;
water color (mg Pt dm−3) – a comparative method, using the Platinum-Cobalt Reference Standards;
photosynthetically active radiation (PAR, in %) – Licor LI–250 Light Meter, % of the light reaching the water surface;
total nitrogen – Merck Spectroquant Cuvette Test;
total phosphorus – microwave mineralization in the presence of nitric and sulfuric acids, followed by the determination of total phosphorus using the Merck Spectroquant test for phosphate determination.
The obtained data were entered into a spreadsheet allowing also for species abundance in plant samples and data describing the environmental factors in the depth zones. For plant samples, species diversity was calculated using the
Relationships between the relative abundance of a given species in the vegetation and pH of water near the sediment were determined by the Kruskal–Wallis non-parametric test (Sokal & Rohlf 2012). The frequency of species was calculated from the quotient of their occurrence in the samples. The relationship between the frequency and abundance of plants and the environmental factors was determined by multiple regression and canonical analysis (Gittins 1985; ter Braak 1987). The data included in the analyses were standardized in CANOCO 4.5 (ter Braak & Šmilauer 2002).
Detrended correspondence analysis (DCA) showed that all data on plant abundance and water properties were linear in their structure and represented only a part of the Gaussian distribution (eigenvalues < 2 SD), which required the use of redundancy analysis RDA (ter Braak & Šmilauer 2002). By using the redundancy analysis (RDA; ter Braak 1986; 1987; 1988), we correlate the frequency and abundance of plants with the physicochemical properties of water. Before the analysis, Spearman’s rank correlation coefficient (R) was calculated to determine strongly correlated variables. And thus, the following parameters were removed from the RDA analysis: water color, as it was strongly correlated with water clarity (
A total of 71 species of macrophytes were identified in 12 702 plant samples collected from 70 soft-water lakes with isoetids: 41 vascular plants, including four isoetids, 19 mosses and 11 stoneworts (Table 1). Twenty-two plant species occurred with a frequency (
Examined species of macrophytes with frequency of over 1%; F – average value of species frequency in the whole pH range and P – average value of species coverage in the whole pH range
Bryophytes | F | A | Stoneworts | F | A | Vascular plants excluding isoetids | F | A | Isoetids | F | A |
---|---|---|---|---|---|---|---|---|---|---|---|
14.1 | 25.8 | 8.4 | 15.5 | 19.0 | 15.2 | 27.4 | 35.0 | ||||
12.0 | 12.7 | 6.1 | 32.5 | 12.4 | 12.8 | 13.5 | 21.8 | ||||
7.9 | 7.6 | 3.6 | 3.2 | 2.1 | 2.3 | 8.6 | 33.3 | ||||
7.0 | 9.7 | 1.7 | 6.1 | 1.8 | 7.1 | 2.9 | 28.8 | ||||
3.6 | 12.0 | 1.1 | 12.3 | 1.6 | 7.6 | - | - | - | |||
1.4 | 8.8 | - | 1.1 | 6.8 | - | - | - | ||||
1.1 | 40.6 | - | - | - | - | - | - | - | - | - |
Mosses are the main constituent of submerged vegetation in acidic lakes (pH 4.0–5.5), with usually one species prevailing in a sample (1.0 ± 1.0 species 0.1 m−2). The moss cover is very dense and may reach more than 90%, especially when pH is low. When pH of water is 4, the moss cover is 30.2±41.0%. In waters with neutral to alkaline pH, the number of moss species and their cover gradually decrease. Moreover, species of the genus
Vascular plants occur throughout the whole pH range (4.0–9.5). In acidic lakes, the number of species is small, on average 0.6±0.4 species 0.1 m−2. However, as pH increases, the number of species gradually increases and reaches the highest level in alkaline lakes (pH>8.5). The number of species in such water bodies increases up to 1.9 ±1.1 species 0.1 m−2. The cover of vascular plants in acidic waters is similar to other plant groups and increases only in alkaline waters, reaching the highest values (40.8±37.8%) at pH=9.
Like other vascular plants, isoetids occur at pH 4.0–9.5, but the largest number of species (0.7 ± 0.8 species 0.1 m−2) occurs in the pH range of 5.5–6.7. One of the isoetid species (
Few stonewort species occur in soft-water lakes (0.6 ± 0.7 species 0.1 m−2). The largest populations of these plants occur at pH 6.5–8.5 (Fig. 3). Stoneworts do not occur in low pH waters, while their relative abundance in submerged vegetation is negligible when pH is higher than 8.5. The abundance of stoneworts is small and they reach the highest cover (11.4 ± 25.6) in waters with neutral or alkaline pH (7–8.5).
Plant species diversity expressed by the
Differences in the Shannon–Wiener index (
Shannon–Wiener index ( |
||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
Mean | 0.17 | 0.41 | 0.21 | 0.18 | 0.32 | 0.23 | 0.37 | 0.31 | 0.31 | 0.32 | 0.24 | 0.40 |
SD | 0.31 | 0.37 | 0.29 | 0.28 | 0.35 | 0.30 | 0.36 | 0.37 | 0.39 | 0.33 | 0.35 | 0.37 |
pH | 4.0 | 4.5 | 5.0 | 5.5 | 6.0 | 6.5 | 7.0 | 7.5 | 8.0 | 8.5 | 9.0 | 9.5 |
4.0 | *** | - | - | *** | * | *** | *** | *** | *** | - | * | |
4.5 | *** | *** | - | *** | - | ** | * | - | * | - | ||
5.0 | - | *** | - | *** | ** | - | ** | - | - | |||
5.5 | *** | - | *** | *** | *** | *** | - | - | ||||
6.0 | * | - | - | - | - | - | - | |||||
6.5 | *** | - | - | - | - | - | ||||||
7.0 | *** | ** | - | ** | - | |||||||
7.5 | - | - | - | - | ||||||||
8.0 | - | - | - | |||||||||
8.5 | - | - | ||||||||||
9.0 | - |
p>0.05; * p<0.05; ** p<0.01; *** p<0.001 (post-hoc Dunn’s test).
The pH of water is a characteristic of the aquatic environment of soft-water lakes with isoetids, which significantly affects the occurrence and abundance of submerged plant species (Fig. 7, Table 3). Other characteristics determining the conditions for the occurrence of plants in lakes, such as PAR, conductivity and water clarity, should also be noted.
Results of RDA analysis of the relationship between plant species and water characteristics in soft-water lakes with isoetids
Axes | 1 | 2 | 3 | 4 | Total variance |
---|---|---|---|---|---|
Eigenvalues | 0.123 | 0.020 | 0.013 | 0.001 | 1.000 |
Species-environment correlations | 0.474 | 0.397 | 0.377 | 0.175 | |
Cumulative percentage variance | |||||
of species data | 12.3 | 14.3 | 15.6 | 15.7 | |
of species-environment relation | 77.5 | 90.3 | 98.4 | 99.2 | |
Sum of all eigenvalues | 1.000 | ||||
Sum of all canonical eigenvalues | 0.158 |
The Monte Carlo Test showed that PAR and pH account for most of the variance (lambda 1 in Marginal Effects is 0.06 and 0.02, respectively), but once the environmental variables are successively included in the model (Conditional Effects), PAR accounts for 0.06 of the variation in the vegetation traits, temperature accounts for 0.05, while pH accounts for 0.02 of the variation. The remaining environmental characteristics, although statistically significant, are less important.
The occurrence of isoetids as a species group is largely affected by factors correlated with the first axis (mainly PAR), but it is the factors correlated with the second axis, mainly pH, that have a decisive impact on the occurrence of individual species.
As far as other species groups are concerned, mosses are associated with high water acidity, while stoneworts are associated with low acidity. The effect of pH on the number of mosses and stoneworts is greater than on the abundance (cover) of their populations. The above analysis indicates that the main course of species turnover along the water pH gradient involves the replacement of mosses by stoneworts, leading to a minor transformation within the group of vascular plants, including isoetids.
Mosses occur with the highest frequency in acidic (pH 4.0–6.5) water (Figs 3, 4).
The first signs of species turnover were observed at low water acidity – pH 6.6–7.5. In plant samples from such lakes, an increase in the frequency of mosses other than those mentioned above, mainly of
The second stage of species replacement was identified in waters with relatively high alkalinity (pH 7.5–9.5). Almost all mosses, as well as isoetids and stoneworts were replaced (Figs 3, 4). Their habitats were occupied by vascular plants. At pH close to neutral,
Soft-water lakes with isoetids (or so-called lobelia lakes) are included in the EU Habitats Directive, entry 3110, in the Natura 2000 network (Kolada et al. 2017). They are of great natural value due to specific characteristics of the aquatic environment and the occurrence of rare plant species from the isoetid group, such as
As a result of environmental changes caused by pH fluctuations, light intensity in water decreases, massive growth of plankton and epiphytic algae becomes more frequent, while the boundaries of areas occupied by sediment-rooted plants are shifted toward the littoral zone (Szmeja 1992; Banas et al. 2012). In these conditions, populations of submerged macrophytes disperse and disintegrate (Szmeja & Bociag 2004; Bociag et al. 2013).
Apart from the reduced light availability, which could have a negative impact on the occurrence and abundance of individual macrophyte species, other environmental parameters are also very important. The research confirms that oxygenation also plays a significant role. A decrease in the oxygen level leads to the withdrawal of all species from their habitat, irrespective of the group to which they belong. It can therefore be said that it is likely that none of the species under study is able to cope with a significant decrease in water oxygenation.
Another important factor is conductivity, which reflects the content of calcium salts in water, varying along with pH. As conductivity increases, vascular plants are favored, particularly those that acquire carbon for photosynthesis from the breakdown of hydrocarbons, thus they cope better in waters with a higher content of calcium, carbonates and pH fluctuations compared to bryophytes, stoneworts, and isoetids, which quickly withdraw under such conditions.
The specificity of the surveyed water bodies is also important. Isoetid lakes have low pH, which results in their low buffering capacity, thus making them less susceptible to adverse external impact compared to lakes with harder water and higher reaction. It could therefore be assumed that pH, which is the product of many chemical parameters, is largely responsible for the resistance of individual water bodies to environmental changes (Szmeja 2006). Furthermore, the trophic status of these lakes is equally important. These are oligotrophic water bodies where – as shown by the research (cf. Fig. 6) – the content of total nitrogen and phosphorus is insignificant in terms of the occurrence and abundance of individual plant species compared to other types of lakes (Banas 2016).
Nevertheless, it is important to bear in mind that natural succession of lakes largely affects underwater conditions and may be a determinant of the occurrence of individual species. It determines environmental conditions (Siraj et al. 2011; Feldmann 2012), causes disruption of interspecific relations (Mitchell & Perrow 1998; Szmeja 2010; Szmeja et al. 2010) and transformation of plant communities (Chmara et al. 2013).
Nearly half of the soft-water lakes with isoetids in Poland are located in forest and are under protection. The remaining lakes are, to a varying extent, affected by human activity, especially in the vicinity of towns and cities. Anthropogenic plant species replacement in soft-water lakes with isoetids is common and should be monitored.