Although the herb layer represents less than 1% of the biomass of temperate forests, this layer may contain up to 90% of the plant species in the forest and can contribute up to 20% of the foliar litter, thus playing an essential role in forest biodiversity and nutrient cycling. The objective of this study was to investigate the differences in cover, species richness (S), and Shannon-Wiener diversity (H) of herbaceous plants a) under tree species associated with contrasting soil nitrogen levels and b) in watersheds that vary in nitrogen deposition, stand age, and nitrate export levels. This study took place in four watersheds at the Fernow Experimental Forest in north-central West Virginia. Watersheds were paired to represent contrasts in nitrogen deposition levels (unfertilized [WS3] versus fertilized [WS7], stand age (young [WS7] versus old [WS13]), and streamwater N levels (high [WS13] versus low [WS10]). Nine plot pairs per watershed were established around the stems of red maple (Acer rubrum) and sugar maple (A. saccharum) with a total of 72 plots in the study. Within each plot, herbaceous species composition and cover were assessed in four circular 1-m² herb sampling quadrats around the stem of each tree.
Results/Conclusions
Analysis of Variance for each watershed pair showed that overstory tree species was not a significant (P>0.05) main effect for any of the herb characteristics studied. However, there was a main effect of watershed (interpreted as treatment effect: fertilization, stand age) and a watershed × overstory tree species interaction effect (P≤0.05) or trend (P≤0.1) for cover and S for the fertilized vs. unfertilized (young) watershed pair and for the young vs. old watershed pair. Herb cover overall was lower in the fertilized than unfertilized watershed (P=0.07) and it was higher under sugar than red maple with fertilization with the opposite pattern in the unfertilized watershed (P=0.06). The ‘young’ watershed had higher understory cover than the ‘old’ watershed (P<0.0001); cover was higher under red than sugar maple in the young watershed and the opposite in the old WS (P=0.01). The pattern was the same for species richness. Diversity was higher in the unfertilized (P=0.0001) and young watershed in their respective pairs (P=0.015). In sum, herb layer characteristics can be influenced by nearby tree species, but the overstory effect depended on nitrogen deposition level and stand age. Thus, further studies elucidating the role of different herb communities in forest functioning are warranted.