Fig 1.
Location of the Loess Plateau in China (left plate) and the distribution of the 188 sampling sites where in situ measurements of aboveground net primary productivity were taken in the study area (right plate).
Table 1.
Statistical values of aboveground net primary productivity (ANPP) and precipitation-use efficiency (PUE) for the three grassland types: MS, meadow steppe (n = 400); TS, typical steppe (n = 3000); DS, desert steppe (n = 1200).
CV% is the percent coefficient of variation: (Standard deviation/Mean). MAPavg is the average mean annual precipitation (MAP) for the grassland types. Values followed by different lower-case letters are significantly different for different grassland types at P < 0.05.
Fig 2.
Correlation between (a) aboveground net primary productivity (ANPP) and mean annual precipitation (MAP) and that between (b) ANPP and precipitation seasonal distribution [PSD, quantified herein with coefficient of variation (CVmp)] and for each grassland type in the Loess Plateau temperate grassland.
Each data point in the figure represents 14-year average value for 2000–2013. R2 is the coefficient of determinant of the regression functions between MAP, CVmp and ANPP for each grassland type.
Fig 3.
Correlations between (a) precipitation use efficiency (PUE) and mean annual precipitation (MAP) and that between (b) PUE and leaf area index (LAI) for the entire Loess Plateau temperate grasslands.
Fig 4.
Correlation between precipitation use efficiency (PUE) and precipitation seasonal distribution (PSD, quantified herein with CVmp) for each grassland type and for the entire Loess Plateau temperate grassland.
Fig 5.
Correlation between (a) aboveground net primary productivity (ANPP) and mean annual precipitation (MAP) and that separately between (b) precipitation-use efficiency (PUE) and MAP and also soil water storage (SWS) at various soil depths at the 101 soil water observation sites in the Loess Plateau grassland.