Nutrient availability determines vegetation patterns and ecological functioning of intact groundwater-fed peatlands (fens). Bryophytes, commonly referred to as ‘brown mosses’, dominate calcareous fens (rich fens), are an integral part of their unique biodiversity and contribute significantly to peat formation and carbon sequestration. Brown mosses are replaced by vascular plants as nutrient availability increases. The decline of brown mosses may either be due to their physiological intolerance of high nutrient levels or to them being outcompeted by vascular plants. We aimed to distinguish between these two hypotheses by investigating whether the ecological optima reflect the physiological optima of brown mosses. Eight brown moss species, common in calcareous fens of the northern hemisphere, were grown under a gradient of nitrogen and phosphorus availability. Biomass increment, chlorophyll content and biomass nutrient concentration were measured. All brown moss species studied achieved the highest growth rates and chlorophyll contents when grown under conditions of nitrogen and phosphorus concentrations higher than those where they occur naturally at the highest frequency. Two of the species showed no growth saturation even at the highest levels of nutrient availability, while the others appeared potassium-limited at the highest N and P levels. Brown mosses dominate natural fens at the lower edge of their physiological optimum in terms of nutrient availability, i.e., their realized niche is much narrower than the fundamental one. Based on the literature, we argue that it is the competition for light with vascular plants which limits the occurrence of brown mosses in nutrient-rich habitats and prevents them from occupying their entire fundamental niche.

Analysed species

We chose eight brown moss species that dominate calcareous fens in temperate Europe: Aulacomnium palustre (Hedw.) Schwägr., Calliergon giganteum (Schimp.) Kindb., Calliergonella cuspidata (Hedw.) Loeske, Campylium stellatum (Hedw.) Lange & C. E. O. Jensen, Climacium dendroides (Hedw.) F. Weber & D. Mohr, Hamatocaulis vernicosus (Mitt.) Hedenäs, Scorpidium cossonii (Schimp.) L.E.Anderson, H.A.Crum & W.R.Buck and Tomentypnum nitens (Hedw.) Loeske. These species occupy different microhabitats along gradients of wetness, nutrient availability and productivity under field conditions (de Mars et al.1996, Štechová et al. 2008, Jabłońska et al. 2011, Kooijman 2012, Pawlikowski et al. 2013, Table 1.).

Sampling site

Bryophyte samples were collected in December 2017 from calcareous fens of the upper Biebrza valley, north-eastern Poland (between Ostrowie 53.71979 °N 23.30223 °E and Szuszalewo 53.71355 °N 23.35722 °E). The vascular plant annual production at the sampling sites was 120 – 315 g m-2 and total vascular plant cover was ca. 60%. The moss annual production was 372 – 440 g m-2 y-1 and exceeded the production of vascular plants (own unpublished data). Total moss cover was relatively high (c. 95%). Bryophytes were accompanied by small sedges such as Carex panicea, C. limosa, C. rostrata, and C. lasiocarpa, some dicotyledonous plants, such as Menyanthes trifoliata and Potentilla palustris, and shrub birch Betula humilis. The average water level measured by automatic loggers (logging interval 1 hour) in 9 piezometers located in the proximity of the collection sites between 15/09/2017 and 15/09/2018 was 1.1 cm below the surface (minimum 12 cm below the surface, maximum 6.9 cm above the surface) (own unpublished data). Mean concentrations of the main elements in water samples collected in summer 2017 at a depth of 0 – 5 cm below the moss carpet were (mg·L-1): 0.05 N-NH4+, 0.025 N-NO3-, 0.5 P-PO43-, 0.6 K+, 7.9 S-SO42-, 55.8 Ca2+, 13.5 Mg2+, 7.3 total Fe, 6.1 Na2+ and pH was 6.5 – 6.7.

Preparation of bryophyte material

We sampled 10 – 15 moss patches, each about 10 cm in diameter, from every species studied. Samples were transported to the laboratory, watered with distilled water and
stored in glass bowls in a cooling room under constant light conditions, with a photosynthetic photon flux density of around 2 μmol m−2s−1, at about 4 °C for around 2 months.
Before onset of the experiment, we moved mosses to the greenhouse for 14 days to allow them to acclimate to the new conditions. After the acclimatisation period, the plants did
not show any symptoms of stress, such as tip drying or losing colour. We washed the mosses with distilled water and cut off the top 3 cm of their shoots. We weighed the fresh mass of each sample after blotting it twice with paper towels in a standardized way and recalculated it to dry mass (see Supplement S1 for calibration of the dry mass of bryophytes at the starting point). The experimental units were colony fragments corresponding to about 3 g dry mass per single species placed in stainless-steel baskets (3 cm x 3 cm x 4.5 cm). The density of shoots in the basket was similar to that under natural conditions (based on visual assessments). Each basket was placed in a jar (540 ml, Supplement S2) filled with an experimental nutrient solution. At the beginning of the experiment, the mosses were submerged in the solution up to about 1 cm below their tip.

Experimental design

The experimental N and P gradient ranged from low concentrations, typical for natural fens dominated by brown mosses (Jabłońska et al. 2011), to high values, corresponding to substantial N and P enrichment due to eutrophication in agricultural landscapes (Table 2.). A. palustre, H. vernicosus, S. cossoni, T. nitens were cultivated in nine levels of nutrient availability, C. cuspidata, C. stellatum, C. dendroides in eight levels (without the 9th level), whereas C. giganteum was cultivated in four nutrient levels (2nd, 4th, 6th, and 8th) because of shortage of plant material for these species. Each experimental unit contained one basket in one jar with one moss species in one type of nutrient treatment. We placed experimental units of all species in a box (the experimental block). There were five blocks for each nutrient level as replicates. We covered the plastic boxes with glass to avoid fungal contamination and to assure constant high moisture, but we left a 1 cm gap to enable airflow (Supplement S2).

As a nutrient solution, we used bottled mineral water (Muszyna Cechini), which is naturally rich in bicarbonates – 1314 mg·L-1. The water is medium saturated with CO2 (1500 – 4000 mg·L-1). The concentrations of the main elements, after mixing 1 part of Muszyna water with 3 parts of distilled water, corresponded to the average element concentrations measured in the groundwater of well-preserved calcareous fens. The concentrations of elements in solution were (mg·L-1): K 1.24, S 4.56, Ca 56.25, Mg 13.35, Na 11,05, Fe 0.14, Mn 0.05, B 0.03, Zn 0.0003, Cu 0.0003, Mo 0.00005, and Co 0.00008. Measurements were performed with Inductively Coupled Plasma Mass Spectrometer ICP MS (Perkin Elmer NexION 300D, USA). We did not add extra K to the solution because K availability is usually stable along natural and anthropogenic nutrient gradients in calcareous fens, as it is rather related to the local climate or hydrogeology (Venterink et al. 2002). Thus, K concentration was constant in all NP treatments. N and P were added as NH4NO3 and NH4H2PO4 taking into account the initial N concentration (0.075 mg N·L-1) present in the solution at each level (from ammonium (N-NH4+) and nitrates (N-NO3) dissolved in the mineral water; measured − with a SAN++ continuous flow analyser (Skalar, Netherlands)). A logarithmic scale of N and P concentrations was applied. N and P concentrations at the first and second levels corresponded to those measured in groundwater of well-preserved calcareous fens in north-eastern Poland (mean concentrations in Biebrza and Rospuda fens were (mg·L-1): 0.05 N-NH4+, 0.025 N-NO3-, 0.5 P-PO43-, 0.6 K+; own data and Jabłońska et al. (2011)), where the bryophyte species studied occur with high abundances. Subsequent levels of nutrient concentrations corresponded to more eutrophic calcareous fens (e.g. Lenz and Wild 2001, Kieckbusch and Schrautzer 2007) up to almost 25 mg N L-1 in the solution at the 9th level typical for highly nutrient-enriched wetlands (Emsens et al. 2016). The ratio between N and P concentrations in the water was kept constant at about 1.6 (between 1.59 and 1.67) because such a ratio is a median for well-preserved calcareous fens in north-eastern Poland (own data) and indicates N-limitation. The solution with added N and P was changed every 3 days during the experiment and mosses were additionally sprayed with distilled water to avoid shoot drying. We mixed the solution in a 20 L canister and poured it into jars. The pH ranged from 6.5 at the beginning to 7.9 after each 3-day period, because the solution with Muszyna water contained additional free CO2, which was gradually released from the water. By replacing the solutions, we avoided the problem of calcium precipitation (c.f. Vicherová et al. 2015). We avoided possible systematic environmental effects within each block by moving the blocks every 3 days during the experimental period.

The experiment lasted 54 days from 12 February 2018 to 06 April 2018. The average greenhouse temperature was about 10 °C in February and about 15 °C in March and April. The mosses grew in daylight supported by additional illumination with white LED lamps in 16h/8h light/dark cycles with a photosynthetic photon flux density of 196 μmol m−2s−1, which is recommended light regime for bryophyte growth (Duckett et al. 2004).

Growth measurement

At the end of the experiment, we oven-dried each sample to constant weight (65 °C, 24 h) and measured the dry mass on an analytical balance (KERN 410-11 Electronic Analytical Balance 41/ resolution: 0.0001g). We calculated the relative biomass increment (RBI) of each moss sample by dividing the dry mass gain at the end of the experiment (dry mass at the end of the experiment minus the initial dry mass of the sample) by the initial dry mass (recalculated from the fresh mass as described in Supplement S1). Any positive number indicates growth, any negative number indicates a loss of biomass over time. 

Chlorophyll content

Using a CCM-300 chlorophyll meter (Opti-Sciences, Inc., Hudson, NH, USA), we measured the ratio of chlorophyll fluorescence at 735 nm (far-red) and that at 700 nm (red) (chlorophyll fluorescence ratio F735/F700) which correlated to the chlorophyll content in vascular plants (Gitelson et al. 1999) and used it as a proxy for chlorophyll content in moss tissue.

N, P, K in bryophyte biomass

From each bryophyte sample, we cut off tops that grew during the experiment (i.e. above the steel basket edge), froze the samples in liquid nitrogen and ground them in a mortar to a homogenous powder. N content was determined by the dry combustion method (Thermo Scientific Flash2000 CHNS/O Analyzer, USA). For P and K analysis, samples were digested in 65% HNO3 using a Speedwave 4 apparatus (Berghof, Germany), and then P content was measured using a SAN++ continuous flow analyzer (Skalar, Nederland), and K content was measured using a flame atomic absorption spectrometer (ContrAA700, Analytik Jena AG, Germany).

Microsoft Excel