Questions
How have species composition and diversity in the herb and bryophyte layer of spruce forest communities developed on unlimed plots and on plots with varying liming intensity under the conditions of decreasing sulphur (S) deposition and unabated high nitrogen (N) deposition?

Location
Spruce stands in the Erzgebirge, East Germany.

Methods
We repeated herb and bryophyte layer surveys on 36 quasi-permanent plots after approx. 23 years (1989/1991 vs 2012/2014). Over this period, the study area experienced a sharp decline in S deposition and repeated liming of the spruce stands. Our plots represent a gradient of liming from zero to four applications. We analyse the change in species composition and species diversity depending on the liming intensity using Ellenberg indicator values (EIV), correspondence analysis, indicator species analysis and diversity parameters (alpha diversity, Lennon similarity).

Results
On the unlimed plots, the species composition of the herb layer remained unchanged over the study period, despite the decrease in the coverage of Calamagrostis villosa and Deschampsia flexuosa and in the total coverage of the herb layer. In contrast, the bryophyte layer showed a significant increase in coverage and species richness with the decrease in S deposition. Liming led to a sharp increase in species richness in the herb layer, due to the appearance of disturbance indicators. The mean EIVs for acidity and nitrogen (EIV-R and EIV-N) also increased significantly. All observed effects increased with increasing liming intensity. We also noticed an increase in species richness over time in the bryophyte layer on the limed plots. However, this was less pronounced than on the unlimed plots, and decreased with increasing liming intensity. This was because the appearance of disturbance indicators or of typical species for the Calamagrostio villosae-Piceetum was partially compensated for by the disappearance of acid indicators as a result of liming. The species turnover was thus much higher in the bryophyte layer than in the herb layer.

Conclusions
In contrast to the herb layer, the bryophyte layer proves to be a sensitive indicator of changes in air pollution. The effects of greatly reduced S deposition on the bryophyte layer outweigh those of continuous N deposition in the study area. Liming acts as a disturbance in both the herb layer and the bryophyte layer. This influence increases with increasing liming intensity. Our results do not support a general application of liming in large, contiguous forest areas. Rather, this measure should be applied in a differentiated manner