Effects of broad-leaved grass inhibitors and nitrogen fertilizer on seed production Elymus nutans in alpine meadow of the Qinghai-Tibet Plateau

Lu Xin Juan Qi ^* Junhu Su Liu Yanjun Zhang Aolong

• Gansu Agricultural University, Lanzhou, China

The alpine meadows of the Tibetan Plateau play a crucial role in the grassland ecosystem. However, due to the rapid growth and strong competitiveness of broad-leaved grasses, the nutritional resources and living space available for Gramineae species are severely restricted in this region. Broad-leaved grasses and noxious weeds have evolved into dominant population, severely limiting grassland production in alpine meadows. A shortage of premium seeds limits grassland ecosystem restoration efforts. Elymus nutans is regarded as a pioneer plant for restoring degraded grassland into meadows dominated by grasses, and for developing cultivated grassland in the Tibetan Plateau region, and the demand for native seeds of E. nutans is increasing. Therefore, this study investigated the effect of combinations of four levels of a broad-leaved grass inhibitor (0, 0.9, 1.5, and 2.1 kg·hm-2 crossed with four levels of nitrogen fertilizer (0, 75, 150, and 225 kg·hm-2) on seed production of E. nutans in Gannan alpine meadow of the Qinghai-Tibet Plateau. We observed that the grass inhibitor significantly (p < 0.05) influenced on fertile tillers (FT), spikelets per fertile tiller (SFT), seeds per spikelet (SS) and panicle length (PL), but not florets per spikelet (FS) (p = 0.145). Nitrogen fertilizer significantly influenced on FT, FS, SS, and PL (p < 0.001), but not SFT (p = 0.068). The interaction of the grass inhibitor and nitrogen fertilizer had no significant effect on any of these seed yield components (p > 0.05). Both the grass inhibitor and nitrogen fertilizer significantly influenced all indicators of seed production (p < 0.001), increasing their values in a dose-dependent manner. Moreover, their interaction proved significant for all indicators (p < 0.001), except for actual seed yield (p > 0.05), demonstrating their synergistic effects. The maximum thousand seed weight (4.66 g) and actual seed yield (365 kg·hm-2) were observed at the highest doss of 2.1 kg·hm-2 of grass inhibitor and 225 kg·hm-2 of nitrogen fertilizer, which were 1.85-fold and 2.94-fold of the control, respectively. Furthermore, significantly positive correlations were observed among seed yield and all yield components. Pathway analysis showed that FT made significant direct contributions to the seed yield.