Root biomass along subtropical to alpine gradients: global implication from Tibetan transect studies

Much uncertainty in estimating root biomass density (RBD, root mass per unit area) of all roots regionally exists because of methodological difficulties and little knowledge about the effects of biotic and abiotic factors on the magnitude and distribution pattern of RBD. In this study, we collected field data of RBD from 22 sites along the Tibetan Alpine Vegetation Transects executed with the same sampling method that covered a relatively undisturbed vegetation gradient from subtropical forests to alpine vegetation. Our field data indicated that RBD significantly decreased with increasing altitudes (r(2) = 0.60, P < 0.001) but had low or non-robust correlations with above.-round biomass density (r(2) = 0.10-0.34), suggesting that RBD can be predicted without reference to shoot biomass. The transect data further revealed that temperature and/or precipitation were likely the major limiting factors for geographical distribution patterns of RBD. The relationships could be expressed as logistic function with a maximum RBD of 200 Mg/ha (12 = 0.59-0.65, P < 0.001). A simple empirical model was developed from the logistic regressions and then globally tested against data for 295 field plots of undisturbed to semi-disturbed vegetarion ranging from the boreal zone to the tropics. In general, the model explained 80% of the RBD variation for 30 field plots along the North-South Transect of Eastern China (r(2) = 0.80, P < 0.0001) and less than half of the variation in the global data-set (r(2) = 0.45: P < 0.0001). The model predictions were strong for temperate evergreen forests, temperate/alpine shrubs and grasslands, boreal tundra, and Mediterranean deserts. Such a global scaling exercise revealed the global distribution pattern of RBD broadly over a range of major biomes, suggesting the possibility to develop a new method for large-scale estimation of root biomass. (C) 2004 Elsevier B.V. All rights reserved