Initial responses of grass litter tissue chemistry and N:P stoichiometry to varied N and P input rates and ratios in Inner Mongolia

•Tissue N and P concentrations respond strongly to fertilization.•Tissue K, Cu and Na concentrations respond weakly to fertilization.•Culm N, P, K, and Na chemistry often responded more strongly than that of leave.•Tissue chemistry of S. krylovii respond more weakly than that of L. chinensis.•Tissue N and P become more decoupled at greater fertilization levels and N:P ratio. Anthropogenic nitrogen (N) and phosphorus (P) inputs can alter the stoichiometry of senesced plant tissues, a key trait controlling nutrient cycling. However, it is unclear how fertilization rate affects plant litter tissue chemistry under varied N:P supply ratios. In a 2-year study, we investigated the effects of N and P supply rates at three N:P input ratios (4:1, 16:1, and 60:1) on the chemical constitution and N:P stoichiometry of the litter of two grasses: Leymus chinensis and Stipa krylovii. We further evaluated the differential responses of chemical constitution and N:P stoichiometry in leaf and culm litter of L. chinensis. Combined N and P fertilization increased soil acidity and plant-available N, but decreased plant-available P, especially when fertilization occurred at N:P ratio=60:1. Litter N and P concentrations showed positive response to N and P inputs, and N concentration increased with fertilization rate under N:P ratio=4:1, but P concentration decreased under N:P ratio=60:1. Furthermore, we found stronger responses of N and P in L. chinensis and culms than in S. krylovii and leaves. Stoichiometric responses became more positive with increasing N and P fertilization level at each ratio. Nitrogen and P inputs also significantly improved potassium, copper, and sodium concentrations in senesced shoots independent of fertilization rates except for sodium at N:P ratio=16:1, which had weaker responses in L. chinensis and leaves than in S. krylovii and culms. The effects of N and P inputs on other elements were primarily influenced by species and organs, but were also idiosyncratically affected by input levels at each ratio. These results indicate that decreasing evenness of N and P inputs may have increasingly severe non-linear impacts on nutrient cycling and that these impacts will be greater in L. chinensis-dominated ecosystems compared to those dominated by S. krylovii.