Systat

Normality was tested with the Shapiro-Wilk test. Data that did not significantly violate normality were subjected to two-way analysis of variance (ANOVA) for repeated measures and post-hoc Tukey test for pairwise comparisons. Data that failed the Shapiro-Wilk test were analyzed using the Kruskal-Wallis one-way ANOVA on ranks followed by post-hoc pairwise comparisons using Dunn's method. Lines of best fit were generated by ordinary least products regression analysis. Analysis of covariance was used to determine whether these relationships differed according to the intervention (hyperoxia or hemodilution). A two-tailed P \ 0.05 was considered statistically significant. Analyses were performed using either Sigmaplot or Systat software (Systat Software Inc., San Jose, CA, USA).

Statistical analyses were performed using SYSTAT (version 13, SYSTAT Software, Chicago, IL). First, normality was assessed using the Shapiro-Wilk test (51) (link). Data that did not violate normality are presented as means Ϯ SE, whereas data that violated normality are presented as medians (25th percentile, 75th percentile). Dichotomous comparisons were made using either Student's t-test with Welch's correction or the Mann-Whitney U-test. Variables measured at multiple time points were analyzed using repeated-measures ANOVA, with P values conservatively adjusted using the Greenhouse-Geisser method (32) (link). Two-sided P values of Յ0.05 were considered statistically significant.

Statistical analyses were performed with the software R (version 2.15.2; R Development Core Team 2012) and the software SYSTAT (version 12; SYSTAT Software Inc., San Jose, CA). Normality of the data was tested using the Shapiro-Wilk test. Methane emissions, NEE and average physico-chemical pore water properties were normally distributed, while the green C. rostrata dry biomass was square-root transformed to approximate a normal distribution. Variation in CH 4 emission, NEE, C. rostrata biomass and selected physico-chemical surface and pore water properties between the different sampling dates were tested using one-way analysis of variance (ANOVA). In addition, Pearson correlation with bootstrapping (0.95 confidence interval) and Bonferroni probability test was used to assess the relationship between CH 4 emission or NEE and C. rostrata biomass or selected physico-chemical pore water properties. Spearman rank correlation analyses were performed to determine trends along the depth profile of the physico-chemical surface and pore water properties using mean values of the replicate samples taken each depth. Correlations were considered statistically significant for p<0.05.