Modelling fine root biomass of boreal tree stands using site and stand variables

Quantification of fine root biomass is needed for estimating the role of roots as carbon (C) stores and sources of C input into the soil, as well as for modelling of forest nutrient cycling. Due to the laborious nature of root biomass determinations, there is a need to develop indirect methods that would allow fine root biomass to be estimated using data on easily measurable stand and site variables. We developed models for estimating tree fine root (diameter ⩽ 2 mm) biomass of boreal forests by compiling data from 95 Finnish forest stands (55 on upland soils and 40 on drained peatlands). Stand basal area predicted fine root biomass (R2-adj. = 0.35, p < 0.001) better than any other stand variable alone. A model that included stand basal area, dominant tree species group [birch/other], soil type [upland soil/drained peatland], temperature sum and the interaction of soil type and temperature sum accounted for 46% of the variation in fine root biomass. Stand basal area, the C:N ratio of the organic layer or upper 0-20 cm peat layer and the dominant tree species group together explained 45% of the variation in fine root biomass. Temperature sum correlated with fine root biomass in opposite ways in upland soils and in drained peatlands: fine root biomass increased with decreasing temperature sum in upland soil sites, whereas in peatlands fine root biomass decreased as the temperature sum decreased. Although the models were based on a large dataset representing well the variation in stand age, basal area, site fertility and climatic conditions in northern European boreal forests, half of the variation in fine root biomass remained unexplained. More comprehensive fine root biomass sampling in relation to spatial and temporal variation in the forthcoming studies, for example in connection with monitoring campaigns, and additional predictors such as soil texture, successional stage and the fine root biomass of ground vegetation might improve the predictive power of fine root biomass models.