Effects of litter addition and removal on seed germination and seedling recruitment of grass and forb functional groups in a semiarid grassland

Introduction Climate and land-use changes have substantially altered litter inputs, which can affect seedling recruitment through various biotic and abiotic factors. However, the mechanisms underlying the effects of litter alterations on seed germination and seedling recruitment in different plant functional groups remain unclear. Methods We conducted two complementary pot experiments in a semiarid grassland in northern China with six blocks and three litter treatments: removal (0 g m -2 ), control (90 g m -2 ), and addition (180 g m -2 ). We measured the germination rate, seedling density of three grass and three forb species, soil properties and photosynthetically active radiation. Results Litter removal significantly decreased the germination rate of grasses, forbs, and total species by 11.1%, 5.4%, and 8.3%, respectively, and reduced the seedling density of grasses and total species by 47.1% and 42.4%, respectively. Litter addition significantly increased the germination rate of grasses by 13.4%, but decreased the germination rate of forbs by 4.3% and the seedling density of forbs by 33.1%. Across species, germination rate and seedling density were positively correlated with seed mass (R² = 0.59 and 0.19, respectively). Forb seedling density was significantly and positively associated with forb germination rate, whereas grass seedling density was not directly related to grass germination rate, and structural equation models indicated that soil temperature remained a significant pathway for grasses. Discussion These findings indicate that both litter removal and litter addition can reduce seedling recruitment, but the pathways differ between grasses and forbs. Grassland management should therefore avoid both excessive litter accumulation and complete litter removal to sustain seedling recruitment under future global change.