Supplemental material
Table S1. Inflorescence shoot frequencies in the intensely clipped plant patches in each of the N addition treatments that were used to test for impacts of apical meristem removal on subsequent shoot production (twenty 5 x 5 cm patches; 4 clipped subplot patches in each of the 5 N addition treatment replicate plots). ‘Apex intact’ values indicate the number of sampled plant patches in which there were no apical shoots within the clipped area, but there was at least one reproductive shoot adjacent to the designated patch within the same E. vaginatum clump (which therefore remained intact after the clipping treatment). ‘Apex removed’ values indicate the number of sampled plant patches containing an apical meristematic shoot that was removed during clipping. Statistical analysis of the data was conducted on the average of the 4 intensely clipped plant patches within each plot to enable even sample sizes (i.e. n=5 per N addition treatment).
Clipping treatment / N addition treatment / First year (2010) / Second year (2011)Apex intact / Apex removed / Apex intact / Apex removed
Intense / Control / 15 / 5 / 15 / 5
Intense / Low N / 17 / 3 / 17 / 3
Intense / High N / 15 / 5 / 14 / 6
Table S2. Inflorescence biomass of E. vaginatum in response to high level N and P additions as part of another study in the same vegetation type (mesic birch hummock tundra) at our site (see Zamin et al 2014 for details). Data were collected in 2011 following 7 years of fertilization. Means of the raw data (g per 40 x 40 cm2; n = 5), are followed by statistical test results from a two-way ANOVA of reciprocal square-transformed data.
Mean / S.E.Control / 0.015 / 0.010
N addition / 0.004 / 0.004
P addition / 0.048 / 0.037
N+P addition / 0.838 / 0.429
F1,16 / P
N effect / 7.0 / 0.018
P effect / 12.3 / 0.003
N x P interaction / 8.4 / 0.010
4
Fig. S1. Map showing the distribution of the nitrogen (N) addition treatment plots (n = 5 for each fertilisation level) across the research site. Each of these plots (5 x 7 m2) contained a total of 12 patches (i.e. individual clumps or tussocks ranging in area from ~0.0025 – 0.01 m2) of E. vaginatum that were randomly assigned to one of three levels of clipping treatment: control (i.e. no defoliation); simulated moderate graze; and simulated intense graze. The actual clipped treatment sampling area within each patch was 0.0025 m2, and the data from the four patches for each clipping treatment within each plot were averaged to yield a single replicate datum point per clipping treatment per plot.
4
Fig. S2. Effect of clipping intensity and soil N addition treatments on inflorescence production in Eriophorum vaginatum in mesic birch hummock tundra in the same growing season as the clipping treatment (a), and in the following year (b). Error bars are standard errors (n=5).
4