Estimating net primary production of natural grassland and its spatio-temporal distribution in China.

The net primary production (NPP) of grassland largely determines terrestrial carbon (C) sinks, and thus plays an important role in the global C cycle. Comprehensive and sequential classification system of grasslands (CSCS) is a unique vegetation classification system (mainly for grassland) that is dependent on quantitative measurement indices [>0°C annual cumulative temperature (Σθ) and moisture index (K-value)]. Based on the relationship of the quantitative classification of CSCS and grassland NPP, a modified model of Carnegie-Ames-Stanford Approach (CASA) was used to predict the grassland NPP and its temporal and spatial distribution in China from 2004 to 2008. The scatter plot of the estimated NPP and the observed NPP showed that the estimated data can be accepted with correlation coefficient of 0.896 (P<0.05). The average annual NPP of grassland from 2004 to 2008 in China ranged from 443.23 to 554.40 g Cm(-2)yr.(-)(1). The NPP also showed spatial-temporal variations. There existed an increasing trend of NPP from the northwest to southeast due to the zonal distribution of vegetation. From the trend of monthly variations, it can be drawn that the NPP accumulation primarily occurred between April and October. The average NPP over seven months from April to October was 482.19 g Cm(-2), or about 88.78% of the annual total. The spatial-temporal trend suggests the importance of water and thermal regimes in determining the grassland NPP (i.e. water and thermal are key limited factors for the grassland production), which is also confirmed by a cluster analysis. The mean annual NPP and the total annual NPP differed significantly among grassland classes corresponding with different Σθ and K-value. The results demonstrate that the grassland NPP and the classes/super-classes in CSCS achieve the optimum coupling.