Fig. 3. Yearly mean flux-weighted NO₃^– concentration (means and 95% confidence intervals) before and after the dieback of the stand. Significant differences between the intact stand (–1) and the relevant year are marked with asterisks. E, intact stand in the expansion area; –1, intact stand 1 yr before the attack; 0, year of the dieback; 1, 1 yr after the dieback; and so on. Each point was determined from a minimum of 10 suction cups. The scheme of the chronosequence is presented in Table 2. Significance of the differences were tested using a one-way ANOVA. Thamhane's T2 test was used as a post-hoc test because variances were not equal according to Levene statistics. A critical probability level of 0.05 was used to indicate significant differences.