Preparation and Planting
Planting materials were bought from commercial suppliers to ensure uniformity of sources, handling and preparation. Seedlings of Carex acutiformis Ehrh. (pond-sedge), Typha latifolia L. (common cattail), Phragmites australis (Cav.) Trin. ex Steud. (common reed) and Schoenoplectus tabernaemontani (C.C.Gmel.) Palla (softstem bulrush) were bought from re-natur GmbH (Ruhwinkel, Germany). The soil used was an aquatic plant soil which is largely composed of peat (Floragard Wasserpflanzen). Water was taken from the Warnow River that traverses the state of Mecklenburg-Vorpommern, Northeast Germany specifically in Mühlendamm (54⁰5’1.50”N and 12⁰9’5.09”E) for the freshwater and close to the river mouth (Schmarl Dorf, 54⁰8’11.72”N and 12⁰5’20.17”E) for the brackish water. The freshwater used had a salinity of 0.31‰ (EC 0.60 mS cm-1) while the brackish water had a salinity of 9.58‰ (EC=15.36 mS cm-1) and a pH of 8.65 and 9.01, respectively.
Seedlings were planted in 78 cm (length) x 49 cm (width) x 32 cm (height) black storage boxes with 10 cm high cobblestone bed covered with geotextile to allow water permeability. One rubber pipe, with a height and diameter of 24.0 cm and 5.5 cm, respectively, was inserted vertically in one corner of each box for easy water changing during treatment application. Then, the boxes were filled up with 25 kg soil on top of the geotextile.
Experimental set-up and treatment application
The experiment was conducted from April to July 2020 in an open area outside the Institute of Biology, University of Rostock. Fifteen boxes were set up in a Randomized Complete Block Design (RCBD), equally distributed in three rows as replicates. In each row, one box was designated per treatment as follows:
C+ : positive (+) control: permanently brackish water
C- : negative (-) control: permanently freshwater
A2b_2f : alternating 2 days brackish water then 2 days freshwater
A4b_4f : alternating 4 days brackish water then 4 days freshwater
A2b_4f : alternating 2 days brackish water then 4 days freshwater
Henceforth, C+ and C- are referred to as salinity levels while A2b_2f, A4b_4f and A2b_4f are salinity pulses.
Six pots of seedlings for each species, P. australis, T. latifolia, C. acutiformis and S. tabernaemontani, were transplanted in a row, arranged randomly, per box. Treatments were applied for 10 weeks to each designated box after the 3-week establishment period wherein plants were exposed to freshwater under waterlogged conditions. We changed the water by sucking the water out from each box using a rubber hose that was inserted into the pipe. Then, depending on the treatment, either freshwater or brackish water was poured into each box through the pipe up to the surface level to mimic peatlands which are normally water-saturated all year round. Water level was maintained by pouring appropriate water into each box daily.
Light (lux) and air temperature (°C) were monitored during the entire study period using HOBO UA-002-64 pendant temperature/light data loggers (HOBO Pendant® Temp/Light, Onset, Cape Cod, Massachusetts, USA). Salinity (‰), water conductivity (mS cm-1), pH and water temperature (°C) were also monitored daily using HQ40D portable multimeter (Hach Lange GmbH, Berlin, Germany) to ensure similarity among treatment replicates. When salinity differed between treatment replications, e.g. due to heavy rainfall, water was replaced accordingly as soon as possible.
Growth measurement
Five individuals per species per treatment from each of the three replicate boxes were marked for weekly measurement of growth variables including plant height, number of leaves, and leaf length and width. Plant height (cm) was measured from the root collar to the tip of the tallest part of the shoot system using a meter stick. The potential maximum height and maximum growth rates were estimated using the Richards growth model (Richards 1959) following equation 1:

Where, Y = height and t = time, and the 5 parameters are:
A: lower asymptote
K: upper asymptote when c=1: If A=0 & c=1, the K is the carrying capacity
B: growth rate
M: maximum growth rate
V > 0: affects near which asymptote maximum growth occurs
C: typically takes a value of 1, otherwise, the upper asymptote is

Leaf length (cm) of all leaves of each marked plant was measured from the base to the tip of each leaf while the leaf width (cm) was measured at the middle to determine maximum width using a ruler. Leaf area (LA, cm2) was calculated by multiplying the leaf length and width, assuming a nearly rectangular shape. The LA per individual plant was taken as the mean of all leaves while the mean LA (MLA) is the average of all LA per treatment per species.
All plants were harvested after 10 weeks, sorted into species per box, and then the aboveground biomass (stem and leaves) and belowground biomass (BGB, roots) parts were separated. Roots were thoroughly cleaned with tap water. AGB and BGB were determined by oven-drying harvested materials at 70 °C for 48 hours or until constant weight and weighed. Root:shoot ratio (RSR) was calculated by dividing the BGB and AGB dry weights.
Photosynthetic pigments
At the end of the experiment, leaf samples were taken from each leaf spot where photosynthesis measurement was done (see explanation below). These were individually weighed (mg) and placed in a 5 ml tube. Then, 3 ml of N,N-Dimethylformamide (DMF) was added. Samples were stored at 4 °C for 24 hours to extract the photosynthetic pigments. Absorption spectra of the extracts were measured with wavelengths ranging from 350 to 750 nm using the spectrophotometer (UV/VIS spectrometer Lambda 2, PerkinElmer, Waltham, Massachusetts, USA). Chlorophyll a and b, as well as carotenoid contents, were calculated using the following formula (Porra et al 1989) to determine ratios between Chlorophyll a and b (Chl a:b) and Chlorophyll a and carotenoid (Chl a:car).
Chl a (µg · g FM-1) = (A663.8 – A750) 12 – (A646.8 – A750) · 3.11 (3)
Chl b (µg · g FM-1) = (A646.8 – A 750) · 20.78 – (A663.8 – A750) · 4.88 (4)
Car (µg · g FM-1) = ((A480-A470) · 1000 – Chl a · 1.12 – Chl b · 34.07) / 245 (5)
where, Ax is the extinction coefficient at a specific wavelength x (nm)
Chlorophyll fluorescence yield
Three of the marked individuals per species and treatment were selected for the weekly measurement of Chlorophyll fluorescence. A JUNIOR Pulse Amplitude Modulated Chlorophyll Fluorometer (JUNIOR-PAM, Heinz Walz GmbH, Effeltrich, Germany) was used to estimate the quantum yield of photochemical energy conversion in photosystem II (PS II) of the acclimated macrophytes. Measurements were done weekly in the morning (AM, between 4:00-8:00) after dark acclimation at night and under high irradiance at noon (NN, between 11:00-14:00). Time of measurement differed due to the increasing day length from spring to summer. Chlorophyll fluorescence yield was measured by connecting the middle part of the youngest fully developed leaf (3rd or 4th from the youngest leaf) to a magnetic leaf clip, with a 0.5 m long, 1.5 mm diameter light guide with the opposite end connected to the JUNIOR-PAM. For S. tabernaemontani, we measured at the middle part of the stem since it is basically leafless and the stem is its photosynthetic organ. The potential photosynthetic rate, measured as the quantum yield of photochemical energy conversion in photosystem II, was calculated as (Genty et al 1989):

where, Yield = quantum yield of photochemical energy conversion in PS II
Fm’ = maximal fluorescence yield of illuminated sample with all PS II centers closed
F = fluorescence yield measured briefly before application of a saturation pulse
The coefficients of the linear regression analysis of the AM yield measurements (Supplemental Fig. 1) were used to correct the AM yield, eliminating the influence of light and representing the yield of dark acclimated samples. Yield difference (∆Yield) was calculated by subtracting the NN from AM yield of the same measurement day per plant. These ΔYield values were used to determine the photosynthetic efficiency of the plants under the combined effects of salinity and light intensity as abiotic stressors. Light intensity (lux) data was taken as the average of the recorded light intensity from two pendant HOBO-data loggers which was then converted into µmol photons m-2s-1 using a conversion factor of 1.41 that was derived based on (Walsby 1997). Light dose was then taken as the total irradiance received by the plant from morning until the specific time of measurement at noon.
Statistical analysis
We fitted models of individual plant heights based on Richards growth model (Eq. 1) with R statistical software Version 4 (R Core Team 2020) and the package BB (Varadhan and Gilbert 2009). Statistical analyses were done using SPSS Version 27 (IBM SPSS Inc., Chicago, Illinois, USA). One-way analysis of variance (ANOVA) was used to determine significant differences in maximum height, maximum growth rate, leaf area, root:shoot ratio and pigment ratios between treatments of the same species. Post hoc pairwise multiple comparisons were carried out using Tukey’s Honestly Significant Difference (HSD) against an alpha-level of 0.05. For some datasets (e.g. ΔYield) where requirements for ANOVA could not be met, Kruskal-Wallis test was used followed by Dunn’s pairwise post hoc comparisons when results were significant at 0.05 alpha level.