Species characteristics and diversity-productivity relationships in young natural forests of the eastern United States

Background/Question/Methods

The relationship between tree diversity and productivity in natural forests has been examined by observational studies at a range of temporal and spatial scales. The direction of the relationship varied. Our study focused on the diversity-productivity relationship in young stands that have not experienced tree harvesting since stand establishment. It has already been proven that in grasslands greater diversity causes greater productivity. We studied if aboveground (AG) biomass productivity increases with species richness and diversity in young natural forest stands with consideration of species characteristics, functional groups, and environmental attributes. The data was obtained from the USDA Forest Service’s Forest Inventory and Analysis (FIA) database. There were 1036 plots from the eastern United States measured between 1998 and 2011 and containing only naturally regenerated stands up to 30 years old. All plots experienced no tree harvesting since establishment. The study assessed the effects of stand-level and environmental-level predictor variables on the response variable AG biomass.

Results/Conclusions

There were a total of 159 species with species richness per plot ranging from 1 to 27. The AG dry biomass growth of the stands was between 0.01 and 16.49 (ton ha^-1yr^-1). The five most dominant species were Populus tremuloides, Acer rubrum, Pinus taeda, Liquidambar styraciflua, and Abies balsamea based on highest importance value percent of 12.37, 8.22, 5.5, 4.82, and 4.27, respectively. The estimates of mean AG biomass growth were significantly different for species functional groups (shade intolerant, moderate, or shade tolerant) and for tree type (conifer or hardwood; p<0.001). We found a positive relationship between diversity and AG biomass productivity (p<0.001). The interaction between species richness and relative density (understocked, fully stocked, or overstocked) showed homogeneity of regression slopes at different stand ages (R²=0.65, p<0.001). The interaction between species richness and forest type (conifer, hardwood, or mixed) was a significant effect in the model for AG biomass (p=0.016, R²=0.33). The predictor variables diversity, stand age, and site index were significant in the multiple regression model (R²=0.37, p<0.001), while the predictor variables aspect and slope did not have a significant effect in the same model. Our results show that an increase in species richness and diversity is associated with correspondingly greater productivity in young forests.