Manipulation experiments
Triplicate mesocosms were grown in climate-controlled growth chambers (PGC-7L2/DE, Percival Scientific, USA) at Umeå Plant Science Centre for 8-10 weeks with 18/6-h day/night photoperiods and 70% relative humidity (corresponding to average growing season photoperiods and humidity conditions at Degerö-Stormyr: www.icos-sweden.se) under different CO2 concentrations and WT levels. Two identical chambers were used for incubation at atmospheric CO2concentrations of 280 ppm and 400 ppm, respectively with two different WT levels (hummocks: 0 and -20 cm, lawns: 0 and -7 cm below capitulum) in each chamber. CO2 levels in the growth chambers were regulated automatically, solely by removal. Thus, the CO2 in the chambers derived from the atmosphere. Target WT levels were obtained by placing the 11 L plastic containers in larger plastic storage containers, which were then filled with deionized water to the target WT level (Fig. S1). The WT was re-adjusted every 2 days, resulting in variations of ± 2 cm. Mesocosms incubated at low WT were sprayed with deionized water every 2 days to maintain the high moisture levels that are naturally provided in the mire by precipitation.
The above-mentioned incubations were conducted in three batches with different temperature and light intensity settings. The first two bathes were performed at light intensities of 250 μmol m-2s-1 (at moss surface) and day/night temperatures of either 12˚C/7˚C or 17˚C/12˚C, whereas the last batch was conducted at 500 μmol m-2 s-1 and 17˚C/12˚C. Thus, the first two incubations represent a full 23factorial design with low light intensity as a fixed factor and atmospheric CO2 concentration, WT and temperature as variables, and the last incubation a 22 factorial design, with temperature and light intensity as fixed factors and CO2 concentration and WT as variables. In the following text, the different CO2, WT, temperature and light settings are referred to as low and high levels respectively. The temperature settings were chosen according to typical growth-period conditions in the field (Peichl et al ., 2014). Light intensity settings were chosen in the range of typical Sphagnum light saturation (250-500 µmol m-2 s-1; Harley et al. , 1989; Jauhiainen & Solviola, 1999; Laine et al. , 2011). The response variables measured in the experiments were: photorespiration/photosynthesis ratio (D6S/D6R ratio), moss height increment, biomass production, 13C and15N discrimination, and concentrations of C and N in the plant tissue.