||Proposed links between biodiversity and ecosystem processes have generated intense interest in the linkage between aboveground net primary productivity (ANPP) and soil C storage. Quantity and quality of ANPP largely depend on plant functional groups and management practices. In a context of environmental change (that is, land-use and climate) long-term studies of ANPP and functional groups are gaining interest. However, rapid determination of ANPP and functional groups are often limited in time and money, resulting in less than ideal sampling schemes and replications. Near-infrared reflectance spectroscopy (NIRS) can relieve constraints of labor intensive hand-sorting by providing quick, non-destructive, and quantitative analyses of a range of organic constituents (for example, plant tissues). Here, we investigated the potential of a NIRS method to rapidly predict harvested green aboveground biomass, the proportion of dead material, and simple functional plant traits, necessary to determine ANPP and related ecosystem properties. The issue was investigated for two independent grassland experiments of contrasted long-term field management (high vs. low grazing and N fertilization). Our results show that NIRS analyses are well suited to determine ANPP (12 and 19% error of prediction) and simple plant traits (error 9%) of contrasted treatment of two independent multi-species grasslands. Moreover, we show that calibration may be simplified when compared to commonly used protocols, which offers ecologists enormous analytical power.