Background/Question/Methods Nutrient use efficiency is an integrative measure of ecosystem functioning, and increased nutrient use efficiency with increasing species richness may be an important mechanism for explaining the observed positive relationship between biodiversity and productivity. Experimental evidence, however, is still lacking for understanding the effects of biodiversity on plant nutrient use efficiency. As part of the ongoing large-scale Inner Mongolia Grassland Removal Experiment (IMGRE) on the relationship between biodiversity and ecosystem functioning, this study was designed to test whether the richness and composition of plant functional group would affect several ecosystem processes, using a plant functional group richness gradient generated by removing particular combinations of plant functional groups. The specific measures of ecosystem processes include the net nitrogen mineralization rate, soil mineralized nitrogen concentration (KCl-extractable [NO3-] and [NH4+]), N concentrations, nitrogen resorption, and nitrogen use efficiency of plants at the species, functional group and ecosystem levels. Results/Conclusions In the third year after functional group removal, we found that plant functional group richness was positively correlated with soil net nitrogen mineralization and nitrification rates, but negatively correlated with soil inorganic nitrogen concentrations. The presence of perennial bunch grasses significantly decreased soil inorganic nitrogen concentrations. Since the second year after removal treatment, both plant nitrogen resorption efficiency and use efficiency were positively correlated with plant functional group richness. These results suggest that plants in more functionally diverse mixtures may be more efficient in nutrient use. This finding may indicate an important nutrient retention mechanism for increasing community productivity and stability in this semiarid region.