Search

Knowledge about grassland biomass and its dynamics is critical for studying regional carbon cycles and for the sustainable use of grassland resources. In this study, we investigated the spatio-temporal variation of biomass in the Xilingol grasslands of northern China. Field-based biomass samples and MODIS time series data sets were used to establish two empirical models based on the relationship of the normalized difference vegetation index (NDVI) with above-ground biomass (AGB) as well as that of AGB with below-ground biomass (BGB). We further explored the climatic controls of these variations. Our results showed that the biomass averaged 99.01 Tg (1 Tg=10(12) g) over a total area of 19.6 × 10(4) km(2) and fluctuated with no significant trend from 2001 to 2012. The mean biomass density was 505.4 g/m(2), with 62.6 g/m(2) in AGB and 442.8 g/m(2) in BGB, which generally decreased from northeast to southwest and exhibited a large spatial heterogeneity. The year-to-year AGB pattern was generally consistent with the inter-annual variation in the growing season precipitation (GSP), showing a robust positive correlation (R(2)=0.82, P<0.001), but an opposite coupled pattern was observed with the growing season temperature (GST) (R(2)=0.61, P=0.003). Climatic factors also affected the spatial distribution of AGB, which increased progressively with the GSP gradient (R(2)=0.76, P<0.0001) but decreased with an increasing GST (R(2)=0.70, P<0.0001). An improved moisture index that combined the effects of GST and GSP explained more variation in AGB than did precipitation alone (R(2)=0.81, P<0.0001). The relationship between AGB and GSP could be fit by a power function. This increasing slope of the GSP-AGB relationships along the GSP gradient may be partly explained by the GST-GSP spatial pattern in Xilingol. Our findings suggest that the relationships between climatic factors and AGB may be scale-dependent and that multi-scale studies and sufficient long-term field data are needed to examine the relationships between AGB and climatic factors.